[go: up one dir, main page]

WO2024186885A1 - Systems, methods, and devices for using prognostic indicators of chronic outcome in traumatic brain injury - Google Patents

Systems, methods, and devices for using prognostic indicators of chronic outcome in traumatic brain injury Download PDF

Info

Publication number
WO2024186885A1
WO2024186885A1 PCT/US2024/018646 US2024018646W WO2024186885A1 WO 2024186885 A1 WO2024186885 A1 WO 2024186885A1 US 2024018646 W US2024018646 W US 2024018646W WO 2024186885 A1 WO2024186885 A1 WO 2024186885A1
Authority
WO
WIPO (PCT)
Prior art keywords
mtbi
subject
model
baseline
indication
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/US2024/018646
Other languages
French (fr)
Inventor
Raphael T. GERRATY
Ragini VERMA
Allyson T. GAGE
Lee J. LANCASHIRE
Magali Haas
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Cohen Veterans Bioscience Inc
Original Assignee
Cohen Veterans Bioscience Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Cohen Veterans Bioscience Inc filed Critical Cohen Veterans Bioscience Inc
Publication of WO2024186885A1 publication Critical patent/WO2024186885A1/en
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0033—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room
    • A61B5/004—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
    • A61B5/0042—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part for the brain
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/40—Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4058—Detecting, measuring or recording for evaluating the nervous system for evaluating the central nervous system
    • A61B5/4064—Evaluating the brain
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/40—Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
    • A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/501—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of the head, e.g. neuroimaging or craniography
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Clinical applications
    • A61B8/0808—Clinical applications for diagnosis of the brain
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training

Definitions

  • mTBI Mild Traumatic Brain Injury
  • Traumatic Brain Injury Patients diagnosed with mild Traumatic Brain Injury (mTBI) experience a wide variety of recovery trajectories and chronic symptoms. Despite its prevalence, the long-term outcomes following an mTBI are highly variable and unpredictable. A considerable proportion of patients, estimated between 20-50%, experience persistent or worsening functional impairments years after the injury [5–10]. Chronic psychiatric and functional problems are notably more common in patients with mTBI relative to those with non-head injuries [11–13], encompassing a broad spectrum of symptoms and diagnoses such as Post Traumatic Stress Disorder (PTSD) [14–16], cognitive deficits, insomnia, Major Depression (MDD) [17–20], and Generalized Anxiety (GAD) [21–23].
  • PTSD Post Traumatic Stress Disorder
  • MDD Major Depression
  • GAD Generalized Anxiety
  • the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15.
  • GCS Glasgow Coma Scale
  • One or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a month of incurring the traumatic brain injury.
  • the one or more liquid biopsy-based baseline biomarker results comprises a baseline high-sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject.
  • the one or more liquid biopsy-based baseline biomarker results comprises a neuron-specific enolase (NSE) measurement or a S100B measurement of the mTBI subject.
  • NSE neuron-specific enolase
  • a radiographic modality finding of the mTBI subject is obtained within a month of incurring the traumatic brain injury.
  • the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage. In some embodiments, the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. [0012] In some embodiments, the radiographic modality finding of the mTBI subject consists of (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. DB2/ 47566363.1 3 Attorney Ref.
  • the radiographic modality finding of the mTBI subject comprises a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture.
  • the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury.
  • the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury.
  • An indication is provided, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit.
  • the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit.
  • the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit.
  • the method further comprises obtaining a baseline heart rate, a diastolic blood pressure, a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject.
  • the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline heart rate, the diastolic blood pressure, the respiratory rate, the temperature, the systolic blood pressure, or any combination thereof, of the mTBI subject.
  • the method further comprises obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O 2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject.
  • the indication as to whether the mTBI subject will incur the DB2/ 47566363.1 4 Attorney Ref. No.: 118547-5008-WO chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O 2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject.
  • the providing the indication comprises: inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy- based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit.
  • the model is an Elastic Net model with a multinomial likelihood function.
  • the model is a nonnegative matrix factorization model.
  • the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof.
  • the inputting further comprises inputting one or more covariates into the model.
  • the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof.
  • the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises treating the mTBI subject based on the indication.
  • the treating the mTBI subject based on the indication comprises administering psychotherapy to the mTBI subject.
  • the DB2/ 47566363.1 5 Attorney Ref. No.: 118547-5008-WO treating the mTBI subject based on the indication comprises administering physical therapy to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering speech therapy to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering occupational therapy to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering melatonin or zolipidem to the mTBI subject.
  • the treating the mTBI subject based on the indication comprises administering a selective serotonin reuptake inhibitor to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering methylphenidate or amphetamine salt to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering cyclobenzaprine to the mTBI subject. [0030] In some embodiments, the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises adjusting a treatment regime of the mTBI subject based on the indication.
  • the adjusting the treatment regime comprises increasing a frequency or duration of psychotherapy for the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a frequency or a duration of physical therapy for the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a frequency or a duration of speech therapy for the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a frequency or a duration of occupational therapy for the mTBI subject.
  • the adjusting the treatment regime comprises increasing a dosage of melatonin or zolipidem given to the mTBI subject or increasing a length of time the melatonin or zolipidem is given to the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a dosage of a selective serotonin reuptake inhibitor given to the mTBI subject or increasing a length of time the selective serotonin reuptake inhibitor is given to the mTBI subject.
  • the adjusting the treatment regime comprises increasing a dosage of methylphenidate or amphetamine salt given to the mTBI subject or increasing a length of time the methylphenidate or amphetamine salt is given to the mTBI subject
  • the adjusting the treatment regime comprises DB2/ 47566363.1 6 Attorney Ref. No.: 118547-5008-WO increasing a dosage of cyclobenzaprine give to the mTBI subject or increasing a length of time the cyclobenzaprine is given to the mTBI subject.
  • Figure 1 illustrates an exemplary system topology including a computer system, in accordance with an exemplary embodiment of the present disclosure for determining whether a mild traumatic brain injury (mTBI) subject will incur functional outcome deficit.
  • Figures 2A and 2B illustrate selecting the number of Nonnegative Matrix Factorization (NMF) dimensions for 12 month symptom data.
  • NMF Nonnegative Matrix Factorization
  • FIG. 3A Figures 3A and 3B illustrate present and missing baseline biological measurements.
  • 3A Biological measures collected at the time of injury including vitals, labs, MRI, CT, and hsCRP, neuron-specific enolase (NSE), and S100B levels for all participants included in the symptom clustering.
  • FIG. 4A, 4B, and 4C illustrate a heatmap representing Pearson correlation coefficients between individual scale items for PHQ9, ISI, MPAI, SWLS, BSI, PCL5, and GOS-E assessments in 1377 mTBI patients, in accordance with an embodiment of the present DB2/ 47566363.1 7 Attorney Ref. No.: 118547-5008-WO disclosure.
  • the labels for the horizontal axis share the same numbering as the labels for the vertical axis.
  • FIGS 5A, 5B, and 5C illustrate clustering of patients into chronic symptom sub- types.
  • 5A Heatmap illustrating weights from the NMF basis matrix, representing loadings for each scale item for each cluster.
  • Clusters 1-5 corresponded to PTSD (re-experiencing and avoidance), Life Satisfaction, Depression, Sleep Disturbance, and Functional Outcome Deficits, respectively.
  • 5B tSNE dimensionality reduction of patient scores in the space of scale items. Shading corresponds to cluster assignments and opacity corresponds to assignment certainty, which was operationalized as the negative entropy of normalized NMF coordinates for each subject.
  • 5C Histogram showing the number of patients assigned to each cluster.
  • FIGS 8A, 8B, 8C, 8D, 8E, 8F, 8G, 8H, 8I, and 8J illustrate feature importance for all cluster predictions (cluster 1: 8A and 8B; cluster 2: 8C and 8D; cluster 3: 8E and 8F; cluster 4: 8G and 8H; and cluster 5: 8I and 8J) in accordance with an embodiment of the present disclosure.
  • Feature importance for each measurement included in the model presented in terms of two statistics measured over random samples: the proportion of samples DB2/ 47566363.1 8 Attorney Ref.
  • FIG. 9A, 9B, 9C, 9D, 9E, 9F, and 9G describe determining whether a mild traumatic brain injury (mTBI) subject will incur functional outcome deficit in accordance with the present disclosure in which optional elements are indicated by dashed boxes.
  • mTBI mild traumatic brain injury
  • baseline means soon after incurring a traumatic brain injury.
  • a baseline measurement is taken of a mild traumatic brain injury subject in order to diagnose the subject’s injury.
  • This baseline DB2/ 47566363.1 10 Attorney Ref. No.: 118547-5008-WO measurement can be, for instance, within a day, within a week, within two weeks or within a month of the subject incurring the brain injury.
  • the term “if” may be construed to mean “when” or “upon” or “in response to determining” or “in response to detecting,” depending on the context.
  • the phrase “if it is determined” or “if [a stated condition or event] is detected” may be construed to mean “upon determining” or “in response to determining” or “upon detecting [the stated condition or event]” or “in response to detecting [the stated condition or event],” depending on the context.
  • the term “about” or “approximately” can mean within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which can depend in part on how the value is measured or determined, e.g., the limitations of the measurement system. For example, “about” can mean within 1 or more than 1 standard deviation, per the practice in the art.
  • “About” can mean a range of ⁇ 20%, ⁇ 10%, ⁇ 5%, or ⁇ 1% of a given value. Where particular values are described in the application and claims, unless otherwise stated, the term “about” means within an acceptable error range for the particular value. The term “about” can have the meaning as commonly understood by one of ordinary skill in the art. The term “about” can refer to ⁇ 10%. The term “about” can refer to ⁇ 5%. [0054] As used interchangeably herein, the term “classifier” or “model” refers to a machine learning model. [0055] In some embodiments, a model is an unsupervised learning model. One example of an unsupervised learning model is cluster analysis. In some embodiments, a model includes supervised machine learning.
  • Nonlimiting examples of supervised learning models include, but are not limited to, logistic regression models, neural networks, support vector machines, Naive Bayes models, nearest neighbors models, random forest models, decision trees, boosted trees, multinomial logistic regression models, linear models, linear regression models, Gradient Boosting models, mixture models, hidden Markov models, Gaussian NB models, linear discriminant analysis models, or any combinations thereof.
  • a model is a multinomial classifier algorithm.
  • a model is a 2-stage stochastic gradient descent (SGD) model.
  • a model is a deep neural network (e.g., a deep-and-wide sample-level model). DB2/ 47566363.1 11 Attorney Ref.
  • the model is a neural network (e.g., a convolutional neural network and/or a residual neural network).
  • Neural networks also known as artificial neural networks (ANNs), include convolutional and/or residual neural networks (deep learning models).
  • ANNs artificial neural networks
  • neural networks are machine learning models that are trained to map an input dataset to an output dataset, where the neural network includes an interconnected group of nodes organized into multiple layers of nodes.
  • the neural network architecture includes at least an input layer, one or more hidden layers, and an output layer.
  • the neural network includes any total number of layers, and any number of hidden layers, where the hidden layers function as trainable feature extractors that allow mapping of a set of input data to an output value or set of output values.
  • a deep learning model is a neural network including a plurality of hidden layers, e.g., two or more hidden layers.
  • each layer of the neural network includes a number of nodes (or “neurons”).
  • a node receives input that comes either directly from the input data or the output of nodes in previous layers, and performs a specific operation, e.g., a summation operation.
  • a connection from an input to a node is associated with a parameter (e.g., a weight and/or weighting factor).
  • the node sums up the products of all pairs of inputs, x i , and their associated parameters.
  • the weighted sum is offset with a bias, b.
  • the output of a node or neuron is gated using a threshold or activation function, f, which, in some instances, is a linear or non-linear function.
  • the activation function is, for example, a rectified linear unit (ReLU) activation function, a Leaky ReLU activation function, or other function such as a saturating hyperbolic tangent, identity, binary step, logistic, arcTan, softsign, parametric rectified linear unit, exponential linear unit, softPlus, bent identity, softExponential, Sinusoid, Sine, Gaussian, or sigmoid function, or any combination thereof.
  • the weighting factors, bias values, and threshold values, or other computational parameters of the neural network are “taught” or “learned” in a training phase using one or more sets of training data.
  • the parameters are trained using the input data from a training dataset and a gradient descent or backward propagation method so that the output value(s) that the ANN computes are consistent with the examples included in the training dataset.
  • the parameters are obtained from a back propagation neural network training process.
  • DB2/ 47566363.1 12 Attorney Ref. No.: 118547-5008-WO Any of a variety of neural networks are suitable for use in accordance with the present disclosure. Examples include, but are not limited to, feedforward neural networks, radial basis function networks, recurrent neural networks, residual neural networks, convolutional neural networks, residual convolutional neural networks, and the like, or any combination thereof.
  • a deep neural network model includes an input layer, a plurality of individually parameterized (e.g., weighted) convolutional layers, and an output scorer.
  • the parameters (e.g., weights) of each of the convolutional layers as well as the input layer contribute to the plurality of parameters (e.g., weights) associated with the deep neural network model.
  • at least 50 parameters, at least 100 parameters, at least 1000 parameters, at least 2000 parameters or at least 5000 parameters are associated with the deep neural network model.
  • Neural networks including convolutional neural networks, suitable for use as models are disclosed in, for example, Vincent et al., 2010, “Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion,” J Mach Learn Res 11, pp.3371-3408; Larochelle et al., 2009, “Exploring strategies for training deep neural networks,” J Mach Learn Res 10, pp.1-40; and Hassoun, 1995, Fundamentals of Artificial Neural Networks, Massachusetts Institute of Technology, each of which is hereby incorporated by reference.
  • Additional example neural networks suitable for use as models are disclosed in Duda et al., 2001, Pattern Classification, Second Edition, John Wiley & Sons, Inc., New York; and Hastie et al., 2001, The Elements of Statistical Learning, Springer- Verlag, New York, each of which is hereby incorporated by reference in its entirety.
  • Additional example neural networks suitable for use as models are also described in Draghici, 2003, Data Analysis Tools for DNA Microarrays, Chapman & Hall/CRC; and Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, each of which is hereby incorporated by reference in its entirety.
  • the model is a support vector machine (SVM).
  • SVMs suitable for use as models are described in, for example, Cristianini and Shawe-Taylor, 2000, “An Introduction to Support Vector Machines,” Cambridge University Press, Cambridge; Boser et al., 1992, “A training algorithm for optimal margin classifiers,” in Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory, ACM Press, Pittsburgh, Pa., pp.142-152; Vapnik, 1998, Statistical Learning Theory, Wiley, New York; Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; Duda, Pattern Classification, Second Edition, 2001, John Wiley & Sons, Inc., pp.259, 262-265; and Hastie, 2001, The Elements of Statistical Learning, Springer, New York; and Furey et al., 2000, Bioinformatics 16, 906- 914, each of which is
  • SVMs When used for classification, SVMs separate a given set of binary labeled data with a hyper-plane that is maximally distant from the labeled data. For certain cases in which no linear separation is possible, SVMs work in combination with the technique of ⁇ kernels ⁇ , which automatically realizes a non-linear mapping to a feature space.
  • the hyper-plane found by the SVM in feature space corresponds, in some instances, to a non-linear decision boundary in the input space.
  • the plurality of parameters (e.g., weights) associated with the SVM define the hyper-plane.
  • the hyper-plane is defined by at least 10, at least 20, at least 50, or at least 100 parameters and the SVM model requires a computer to calculate because it cannot be mentally solved.
  • the model is a Naive Bayes model. Na ⁇ ve Bayes models suitable for use as models are disclosed, for example, in Ng et al., 2002, “On discriminative vs. generative classifiers: A comparison of logistic regression and naive Bayes,” Advances in Neural Information Processing Systems, 14, which is hereby incorporated by reference.
  • a Naive Bayes model is any model in a family of “probabilistic models” based on applying Bayes’ theorem with strong (na ⁇ ve) independence assumptions between the features. In some embodiments, they are coupled with Kernel density estimation. See, for example, Hastie et al., 2001, The elements of statistical learning: data DB2/ 47566363.1 14 Attorney Ref. No.: 118547-5008-WO mining, inference, and prediction, eds. Tibshirani and Friedman, Springer, New York, which is hereby incorporated by reference. [0063] Nearest neighbors. In some embodiments, a model is nearest neighbors. In some implementations, nearest neighbor models are memory-based and include no model to be fit.
  • the nearest neighbors For nearest neighbors, given a query point x 0 (a test subject), the k training points x (r) , r, ... , k (here the training subjects) closest in distance to x 0 are identified and then the point x 0 is classified using the k nearest neighbors.
  • the nearest neighbor rule is refined to address issues of unequal class priors, differential misclassification costs, and feature selection. Many of these refinements involve some form of weighted voting for the neighbors.
  • the number of distance calculations needed to solve the k-nearest neighbor model is such that a computer is used to solve the model for a given input because it cannot be mentally performed.
  • the model is a decision tree. Decision trees suitable for use as models are described generally by Duda, 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp.395-396, which is hereby incorporated by reference.
  • Tree-based methods partition the feature space into a set of rectangles, and then fit a model (like a constant) in each one.
  • the decision tree is random forest regression.
  • one specific model is a classification and regression tree (CART).
  • Other specific decision tree models include, but are not limited to, ID3, C4.5, MART, and Random Forests. CART, ID3, and C4.5 are described in Duda, DB2/ 47566363.1 15 Attorney Ref. No.: 118547-5008-WO 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp.396-408 and pp.411- 412, which is hereby incorporated by reference.
  • the decision tree model includes at least 10, at least 20, at least 50, or at least 100 parameters (e.g., weights and/or decisions) and requires a computer to calculate because it cannot be mentally solved.
  • the model uses a regression formula.
  • the regression formula is determined using any type of regression.
  • the regression is logistic regression.
  • the regression is logistic regression with lasso, L2 or elastic net regularization.
  • those extracted features that have a corresponding regression coefficient that fails to satisfy a threshold value are pruned (removed from) consideration.
  • a generalization of the logistic regression model that handles multicategory responses is used as the model. Logistic regression is disclosed in Agresti, An Introduction to Categorical Data Analysis, 1996, Chapter 5, pp.103-144, John Wiley & Son, New York, which is hereby incorporated by reference.
  • the model makes use of a regression model disclosed in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York.
  • the logistic regression model includes at least 10, at least 20, at least 50, at least 100, or at least 1000 parameters (e.g., weights) and requires a computer to calculate because it cannot be mentally solved.
  • Linear discriminant analysis models In some embodiments, linear discriminant analysis (LDA), normal discriminant analysis (NDA), or discriminant function analysis is a generalization of Fisher’s linear discriminant, a method used in statistics, pattern recognition, and machine learning to find a linear combination of features that characterizes or separates two or more classes of objects or events.
  • the resulting combination is used as the model (linear model) in some embodiments of the present disclosure.
  • Mixture model and Hidden Markov model In some embodiments, the model is a mixture model, such as that described in McLachlan et al., Bioinformatics 18(3):413-422, 2002. In some embodiments, in particular, those embodiments including a temporal DB2/ 47566363.1 16 Attorney Ref. No.: 118547-5008-WO component, the model is a hidden Markov model such as described by Schliep et al., 2003, Bioinformatics 19(1):i255-i263. [0069] Clustering. In some embodiments, the model is an unsupervised clustering model.
  • the model is a supervised clustering model.
  • Clustering algorithms suitable for use as models are described, for example, at pages 211-256 of Duda and Hart, Pattern Classification and Scene Analysis, 1973, John Wiley & Sons, Inc., New York, (hereinafter “Duda 1973”) which is hereby incorporated by reference in its entirety.
  • the clustering problem is described as one of finding natural groupings in a dataset. To identify natural groupings, two issues are addressed. First, a way to measure similarity (or dissimilarity) between two samples is determined. This metric (e.g., similarity measure) is used to ensure that the samples in one cluster are more like one another than they are to samples in other clusters.
  • a mechanism for partitioning the data into clusters using the similarity measure is determined.
  • One way to begin a clustering investigation is to define a distance function and to compute the matrix of distances between all pairs of samples in the training set. If distance is a good measure of similarity, then the distance between reference entities in the same cluster is significantly less than the distance between the reference entities in different clusters.
  • clustering does not use a distance metric. For example, in some embodiments, a nonmetric similarity function s(x, x') is used to compare two vectors x and x'.
  • s(x, x') is a symmetric function whose value is large when x and x' are somehow “similar.”
  • clustering techniques contemplated for use in the present disclosure include, but are not limited to, hierarchical clustering (agglomerative clustering using a nearest-neighbor algorithm, farthest-neighbor algorithm, the average linkage algorithm, the centroid algorithm, or the sum-of-squares algorithm), k-means clustering, fuzzy k-means clustering algorithm, and Jarvis-Patrick clustering.
  • the clustering includes unsupervised clustering (e.g., with no preconceived number of clusters and/or no predetermination of cluster assignments).
  • Ensembles of models and boosting In some embodiments, an ensemble (two or more) of models is used.
  • a boosting technique such as AdaBoost is DB2/ 47566363.1 17 Attorney Ref. No.: 118547-5008-WO used in conjunction with many other types of learning algorithms to improve the performance of the model.
  • AdaBoost DB2/ 47566363.1 17 Attorney Ref. No.: 118547-5008-WO used in conjunction with many other types of learning algorithms to improve the performance of the model.
  • the output of any of the models disclosed herein, or their equivalents is combined into a weighted sum that represents the final output of the boosted model.
  • the plurality of outputs from the models is combined using any measure of central tendency known in the art, including but not limited to a mean, median, mode, a weighted mean, weighted median, weighted mode, etc.
  • the plurality of outputs is combined using a voting method.
  • a respective model in the ensemble of models is weighted or unweighted.
  • the term “parameter” refers to any coefficient or, similarly, any value of an internal or external element (e.g., a weight and/or a hyperparameter) in an algorithm, model, regressor, and/or classifier that can affect (e.g., modify, tailor, and/or adjust) one or more inputs, outputs, and/or functions in the algorithm, model, regressor and/or classifier.
  • a parameter refers to any coefficient, weight, and/or hyperparameter that can be used to control, modify, tailor, and/or adjust the behavior, learning, and/or performance of an algorithm, model, regressor, and/or classifier.
  • a parameter is used to increase or decrease the influence of an input (e.g., a feature) to an algorithm, model, regressor, and/or classifier.
  • a parameter is used to increase or decrease the influence of a node (e.g., of a neural network), where the node includes one or more activation functions.
  • Assignment of parameters to specific inputs, outputs, and/or functions is not limited to any one paradigm for a given algorithm, model, regressor, and/or classifier but can be used in any suitable algorithm, model, regressor, and/or classifier architecture for a desired performance.
  • a parameter has a fixed value.
  • a value of a parameter is manually and/or automatically adjustable.
  • a value of a parameter is modified by a validation and/or training process for an algorithm, model, regressor, and/or classifier (e.g., by error minimization and/or backpropagation methods).
  • an algorithm, model, regressor, and/or classifier of the present disclosure includes a plurality of parameters.
  • the plurality of parameters is n parameters, where: n ⁇ 2; n ⁇ 5; n ⁇ 10; n ⁇ 25; n ⁇ 40; n ⁇ 50; n ⁇ 75; n ⁇ 100; n ⁇ 125; n ⁇ 150; n ⁇ 200; n ⁇ 225; n ⁇ 250; n ⁇ 350; n ⁇ 500; n ⁇ 600; n ⁇ 750; n ⁇ 1,000; n ⁇ 2,000; n ⁇ 4,000; n ⁇ 5,000; n ⁇ 7,500; n ⁇ 10,000; n ⁇ 20,000; n ⁇ 40,000; n ⁇ 75,000; n ⁇ 100,000; n ⁇ 200,000; n ⁇ 500,000, n ⁇ 1 x 10 6 , n ⁇ 5 x 10 6 , or n ⁇ 1 x 10 7 .
  • n is between 10,000 and 1 x 10 7 , between 100,000 and 5 x 10 6 , or between 500,000 and 1 x 10 6 .
  • the algorithms, models, regressors, and/or classifier of the present disclosure operate in a k-dimensional space, where k is a positive integer of 5 or greater (e.g., 5, 6, 7, 8, 9, 10, etc.). As such, some embodiments of the algorithms, models, regressors, and/or classifiers of the present disclosure cannot be mentally performed.
  • a computer system 100 is represented as single device that includes all the functionality of the computer system 100, such as one or more processing units (e.g., one or more central processing units and/or one or more graphics processing units).
  • processing units e.g., one or more central processing units and/or one or more graphics processing units.
  • the functionality of the computer system 100 is spread across any number of networked computers and/or reside on each of several networked computers and/or by hosted on one or more virtual machines and/or containers at a remote location accessible across a communications network (e.g., communications network 106 of Figure 1).
  • Figure 1 depicts a block diagram of a distributed computer system (e.g., computer system 100) according to some embodiments of the present disclosure.
  • the computer system 100 at least facilitates communicating one or more instructions for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit.
  • the communication network 106 optionally includes the Internet, one or more local area networks (LANs), one or more wide area networks (WANs), other types of networks, or a combination of such networks.
  • Examples of communication networks 106 include the World Wide Web (WWW), an intranet and/or a wireless network, such as a cellular telephone network, a wireless local area DB2/ 47566363.1 19 Attorney Ref. No.: 118547-5008-WO network (LAN) and/or a metropolitan area network (MAN), and other devices by wireless communication.
  • the wireless communication optionally uses any of a plurality of communications standards, protocols and technologies, including Global System for Mobile Communications (GSM), Enhanced Data GSM Environment (EDGE), high-speed downlink packet access (HSDPA), high-speed uplink packet access (HSUPA), Evolution, Data-Only (EV-DO), HSPA, HSPA+, Dual-Cell HSPA (DC-HSPDA), long term evolution (LTE), near field communication (NFC), wideband code division multiple access (W-CDMA), code division multiple access (CDMA), time division multiple access (TDMA), Bluetooth, Wireless Fidelity (Wi-Fi) (e.g., IEEE 802.11a, IEEE 802.11ac, IEEE 802.11ax, IEEE 802.11b, IEEE 802.11g and/or IEEE 802.11n), voice over Internet Protocol (VoIP), Wi- MAX, a protocol for e-mail (e.g., Internet message access protocol (IMAP) and/or post office protocol (POP)), instant messaging (e.g., extensible messaging
  • the computer system 100 includes one or more processing units (CPUs), such as one or more central processing units (CPUs) and/or one or more graphic processing units (GPUs) 102 a network or other communications interface 104, and a memory 112.
  • the computer system 100 includes a user interface 106.
  • the user interface 106 typically includes a display 108 for presenting media.
  • the display 108 is integrated within the computer systems (e.g., housed in the same chassis as the CPU 102 and the memory 112).
  • the computer system 100 includes one or more input device(s) 110 that allow a subject to interact with the computer system 100.
  • the one or more input devices 110 include a keyboard, a mouse, and/or other input mechanisms.
  • the display 108 includes a touch-sensitive surface (e.g., where display 108 is a touch-sensitive display or computer system 100 includes a touch pad).
  • the computer system 100 presents media to a user through the display 108. Examples of media presented by the display 108 include one or more images, a video, audio (e.g., waveforms of an audio sample), or a combination thereof.
  • the one or more images, the video, the audio, or the combination thereof is DB2/ 47566363.1 20 Attorney Ref.
  • No.: 118547-5008-WO presented by the display 108 through a client application e.g., a client application hosted by the computer system 100, a client application accessible through the computer system 100, etc.
  • the audio is presented through an external device (e.g., speakers, headphones, input/output (I/O) subsystem, etc.) that receives audio information from the computer system 100 and presents audio data based on this audio information.
  • the user interface 106 also includes an audio output device, such as speakers or an audio output for connecting with speakers, earphones, or headphones.
  • the memory 112 includes high-speed random access memory, such as DRAM, SRAM, DDR RAM, or other random access solid state memory devices, and optionally also includes non-volatile memory, such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid state storage devices.
  • the memory 112 may optionally include one or more storage devices remotely located from the CPU(s) 102.
  • the memory 112, or alternatively the non-volatile memory device(s) within memory 112, includes a non-transitory computer readable storage medium. Access to memory 112 by other components of the computer system 100, such as the CPU(s) 102, is, optionally, controlled by a controller.
  • the memory 112 include mass storage that is remotely located with respect to the CPU(s) 102. In other words, some data stored in the memory 112 may in fact be hosted on devices that are external to the computer system 100, but that can be electronically accessed by the computer system 100 over an Internet, intranet, or other form of network 106 or electronic cable using communication interface 104.
  • the memory 112 of the computer system 100 stores: • an operating system 120 (e.g., ANDROID, iOS, DARWIN, RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system such as VxWorks) that includes procedures for handling various basic system services; • an electronic address associated with the computer system 100 that identifies the computer system 100 (e.g., within the communication network 106); • a control module 122 including one or more models 124 for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit; and • optionally, a client application for presenting information (e.g., media) using a display 108 of the computer system 100.
  • an operating system 120 e.g., ANDROID, iOS, DARWIN, RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system such as VxWorks
  • an electronic address associated with the computer system 100 that identifies the computer system 100
  • control module 122 includes one or more models 124 that are configured to perform one or more elements of a method of the present disclosure.
  • Each of the above identified modules and applications correspond to a set of executable instructions for performing one or more functions described above and the methods described in the present disclosure (e.g., the computer-implemented methods and other information processing methods described herein). These modules (e.g., sets of instructions) need not be implemented as separate software programs, procedures or modules, and thus various subsets of these modules are, optionally, combined or otherwise re-arranged in various embodiments of the present disclosure.
  • the memory 112 optionally stores a subset of the modules and data structures identified above. Furthermore, in some embodiments, the memory 112 stores additional modules and data structures not described above. [0083] It should be appreciated that the computer system of Figure 1 is only one example of a computer system 100, and that the computer system 100 optionally has more or fewer components than shown, optionally combines two or more components, or optionally has a different configuration or arrangement of the components. The various components shown in Figure 1 are implemented in hardware, software, firmware, or a combination thereof, including one or more signal processing and/or application specific integrated circuits.
  • TBI mild traumatic brain injury
  • the severity of TBI is graded using the Glasgow Coma Scale (GCS), where mild TBI is defined as GCS > 13, moderate TBI as GCS 9–12, and severe TBI as GCS 3–8. See Teasdale and Jennett, 1974, “Assessment of coma and impaired consciousness: A practical scale,” Lancet 304, 81–84, which is hereby incorporated by reference. DB2/ 47566363.1 22 Attorney Ref. No.: 118547-5008-WO [0086] Referring to block 902, in some embodiments, the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15.
  • GCS Glasgow Coma Scale
  • the GCS is a neurological scale used to assess the level of consciousness of patients.
  • the scale evaluates three aspects of neurological function: eye response, verbal response, and motor response. Each aspect is assessed independently and then combined to give a total score.
  • the total score ranges from 3 to 15, with 3 being the worst possible score (deep coma or brain death) and 15 being the best possible score (fully awake and oriented).
  • the GCS is commonly used in emergency and critical care settings to assess the severity of brain injury and monitor changes in neurological status over time.
  • the mTBI subject has a baseline GCS score in the range of 13- 15 within the month of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15 within a week of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 13-15 within two days of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 13-15 at the time of incurring a traumatic brain injury.
  • GCS Glasgow Coma Scale
  • the mTBI subject has a baseline GCS score in the range of 12- 15 within the month of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 12-15 within a week of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 12-15 within two days of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 12-15 at the time of incurring a traumatic brain injury.
  • the mTBI subject has a baseline GCS score in the range of 11- 15 within the month of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 11-15 within a week of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 11-15 within two days of incurring a traumatic brain injury. In some embodiments, the DB2/ 47566363.1 23 Attorney Ref. No.: 118547-5008-WO mTBI subject has a baseline GCS score in the range of 11-15 at the time of incurring a traumatic brain injury.
  • one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a month of incurring the traumatic brain injury.
  • the one or more liquid biopsy-based baseline biomarker results comprises a baseline high- sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject.
  • hs-CRP is a marker of inflammation in the body and, in some embodiments, is measured through a blood test. For instance, in some embodiments, to obtain the hs-CRP measurement of the mTBI subject, a healthcare provider draws a sample of blood from a vein, usually in the arm, using a needle and syringe. The procedure is similar to other blood tests.
  • the concentration of hs-CRP is then measured in the blood sample in e.g., in units of milligrams per liter (mg/L) of blood. This measurement reflects the amount of hs-CRP present in the bloodstream at the time the sample was collected. See Xu et al., 2021, “High- Sensitivity C-Reactive Protein Is a Prognostic Biomarker of Six-Month Disability after Traumatic Brain Injury: Results from the TRACK-TBI Study,” Journal of Neurotrauma 38, pp.918-927, which is hereby incorporated by reference.
  • the one or more liquid biopsy-based baseline biomarker results comprises a neuron-specific enolase (NSE) measurement or a S100B measurement of the mTBI subject.
  • NSE neuron-specific enolase
  • S100B S100B measurement of the mTBI subject.
  • NSE is an enzyme that is primarily found in neurons, particularly in the central nervous system (brain and spinal cord). It is also present in neuroendocrine cells and certain peripheral nerves. NSE plays a role in glycolysis, the metabolic process that breaks down glucose to produce energy within cells. NSE is used as a biomarker for various neurological conditions, particularly those involving neuronal damage or death. Elevated levels of NSE in the blood or cerebrospinal fluid (CSF) can indicate neuronal injury or damage.
  • CSF cerebrospinal fluid
  • NSE measurement is often used in the diagnosis, prognosis, and monitoring of neurological disorders such as TBI. Measurement of NSE typically involves a blood test or analysis of cerebrospinal fluid.
  • S100B is a protein belonging to the S100 protein family, which consists of calcium- binding proteins found predominantly in cells of the nervous system and some other tissues. S100B, specifically, is primarily expressed by astrocytes, a type of glial cell in the brain, although it is also present in other cells such as melanocytes. S100B is utilized as a DB2/ 47566363.1 24 Attorney Ref. No.: 118547-5008-WO biomarker for traumatic brain injury.
  • the one or more liquid biopsy-based baseline biomarker results are obtained from one or more liquid biopsy samples acquired from the mTBI subject.
  • the one or more liquid biopsy samples are from the blood, whole blood, plasma, serum, urine, cerebrospinal fluid, fecal material, saliva, sweat, tears, pleural fluid, pericardial fluid, and/or the peritoneal fluid of the mTBI subject [0096]
  • a liquid biopsy sample in the one or more liquid biopsy samples has a volume of between 100 ⁇ L and 950 ⁇ L.
  • the liquid biopsy sample has a volume of between 100 ⁇ L and 950 ⁇ L, between 120 ⁇ L and 900 ⁇ L, between 140 ⁇ L and 800 ⁇ L, between 160 ⁇ L and 700 ⁇ L, between 180 ⁇ L and 600 ⁇ L, between 200 ⁇ L and 550 ⁇ L ⁇ or between 220 ⁇ L and 400 ⁇ L.
  • the liquid biopsy sample in the one or more liquid biopsy samples has a volume of between 10 ⁇ L and 300 ⁇ L.
  • the liquid biopsy sample has a volume of between 15 ⁇ L and 275 ⁇ L, between 20 ⁇ L and 250 ⁇ L, between 25 ⁇ L and 225 ⁇ L, between 30 ⁇ L and 250 ⁇ L, between 35 ⁇ L and 225 ⁇ L, between 40 ⁇ L and 200 ⁇ L ⁇ or between 50 ⁇ L and 125 ⁇ L.
  • the liquid biopsy sample in the one or more liquid biopsy samples has a volume of from about 1 mL to about 50 mL.
  • the liquid biopsy sample has a volume of about 1 mL, about 2 mL, about 3 mL, about 4 mL, about 5 mL, about 6 mL, about 7 mL, about 8 mL, about 9 mL, about 10 mL, about 11 mL, about 12 mL, about 13 mL, about 14 mL, about 15 mL, about 16 mL, about 17 mL, about 18 mL, about 19 mL, about 20 mL, or greater.
  • a radiographic modality finding of the mTBI subject is obtained within a month of incurring the traumatic brain injury.
  • MRI magnetic resonance imagining
  • this MRI involves subjecting the mTBI subject to a main magnetic field (B0) (e.g., 1.5 T or 3.0 T), gradient coils providing small magnetic field variation for spatial localization, and radiofrequency proton excitation with DB2/ 47566363.1 25 Attorney Ref. No.: 118547-5008-WO subsequent decay, signal detection, and postprocessing.
  • B0 main magnetic field
  • B0 e.g. 1.5 T or 3.0 T
  • gradient coils providing small magnetic field variation for spatial localization
  • radiofrequency proton excitation with DB2/ 47566363.1 25 Attorney Ref. No.: 118547-5008-WO subsequent decay, signal detection, and postprocessing.
  • fMRI functional MRI
  • DTI diffusion tensor imaging
  • MRP magnetic resonance perfusion
  • DSC Dynamic susceptibility contrast
  • DCE dynamic contrast-enhanced
  • T1 and T2 * signal are due to the T1 and T2 shortening effects of gadolinium, resulting in high T1 and low T2 or T2 * signal, respectively.
  • a third technique, known as arterial spin labelling is a perfusion sequence that does not require intravenous contrast administration. It harnesses the ability of the MRI to selectively label inflowing arterial blood and monitor tissue perfusion. In so doing, the protons in flowing arterial blood act as endogenous contrast to calculate bolus parameters such as cerebral blood flow. See, Petcharunpaisan et al., 2010, “Arterial spin labeling in neuroimaging,” World J.
  • computed tomography is used to obtain the radiographic modality finding. See, Vidhya et al., 2021, “Automated Detection and Screening of Traumatic Brain Injury (TBI) Using Computed Tomography Images: A Comprehensive Review and Future Perspectives,” . Int. J. Environ. Res.
  • the computed tomography is non- contrast computed tomography.
  • the computed tomography is positron emission tomography (PET) or single-photon emission computed tomography (SPECT).
  • PET positron emission tomography
  • SPECT single-photon emission computed tomography
  • the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage (SAH).
  • SAH subarachnoid hemorrhage
  • the radiographic modality finding of SAH is determined using computed tomography.
  • the radiographic modality finding is a binary indication as to whether or not the mTBI subject has a subarachnoid hemorrhage (SAH).
  • SAH subarachnoid hemorrhage
  • the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage (IVH).
  • IVH intraventricular hemorrhage
  • the radiographic modality finding of an IVH is determined using computed tomography. See, for example, Nawabi et al., “Non ⁇ contrast computed tomography features predict intraventricular hemorrhage growth,” European Radiology 33, pp.7807-7817, which is hereby incorporated by reference.
  • the radiographic modality finding is a binary indication as to whether or not the mTBI subject has an IVH.
  • the radiographic modality finding of the mTBI subject consists of (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage.
  • the radiographic modality finding of the mTBI subject comprises a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture.
  • the radiographic modality is computed tomography.
  • An epidural hematoma refers to a medical condition where bleeding occurs between the skull and the outer layer of the membrane covering the brain (dura mater). This condition typically arises due to trauma or injury to the head, leading to the rupture of blood vessels in the epidural space.
  • CT imaging can be used to diagnose epidural hematomas because it provides detailed cross-sectional images of the brain and surrounding structures.
  • an epidural hematoma appears as a lens-shaped or biconvex mass of blood located between the inner surface of the skull and the dura mater. The hematoma may exert pressure on the brain, leading to symptoms such as headache, dizziness, nausea, vomiting, confusion, weakness, or loss of consciousness.
  • a computed tomography (CT) Marshall score is a system used to categorize traumatic brain injury (TBI) based on findings from CT scans of the head. See, for example, Mohammadifard et al., 2008, “Marshall and Rotterdam Computed Tomography scores in predicting early deaths after brain trauma,” Eur J Transl Myol 28(3), pp.265-273, which is hereby incorporated by reference.
  • An edema refers to an accumulation of fluid in the brain tissue.
  • a contusion refers to a bruise on the brain caused by trauma, such as a blow to the head or rapid deceleration injury.
  • a contusion can be discovered, for example, using computed tomography, magnetic resonance imaging, diffusion-weighted imaging, fluid- attenuated inversion recovery (FLAIR) imaging, gradient echo (GRE) imaging, positron emission tomography, and single photon emission computed tomography.
  • a subdural hematoma refers to a type of bleeding that occurs between the layers of tissue surrounding the brain. It typically develops as a result of trauma to the head, causing DB2/ 47566363.1 29 Attorney Ref. No.: 118547-5008-WO blood vessels to rupture and blood to accumulate in the space between the dura mater (the outermost layer covering the brain) and the arachnoid mater (the middle layer). Subdural hematomas can vary in severity and can be classified based on the time elapsed since the initial bleeding. A subacute subdural hematoma indicates that the hematoma is in an intermediate stage of development, typically between 3 days to 3 weeks after the initial bleeding event.
  • the hematoma may appear more organized and may have different radiological characteristics compared to an acute hematoma.
  • a subacute subdural hematoma can be discovered, for example, using computed tomography, magnetic resonance imaging, ultrasound, angiography, positron emission tomography, and single photon emission computed tomography.
  • An extraaxial hematoma refers to a hematoma that occurs outside the brain tissue, while "hematoma" refers to a collection of blood. Extra-axial hematomas can occur in different locations within the skull, typically between the layers of tissue that cover the brain.
  • extra-axial hematomas include epidural hematomas, subdural hematomas, and subarachnoid hemorrhages. Each of these types of hematomas has distinct characteristics and can result from different causes, such as trauma.
  • An extraaxial hematoma can be discovered, for example, using computed tomography, magnetic resonance imaging, ultrasound, angiography, positron emission tomography, and single photon emission computed tomography.
  • CT is particularly effective in detecting skull fractures because it provides high- resolution images of the bones and surrounding soft tissues. It can detect even subtle fractures that may not be visible on conventional X-rays.
  • a Marshall 234 score is derived from the Marshall classification of traumatic brain injury (MCTC), a CT-scan derived metric using only a few features and has been shown to predict outcome in patients with traumatic brain injury. This system was first published in 1992 and remains one of the most commonly used systems for grading acute traumatic brain injury on the basis of CT findings. See, Mahadewa et al., 2018, “Modified Revised Trauma– Marshall score as a proposed tool in predicting the outcome of moderate and severe traumatic brain injury,” Emerg Med.10, pp.135-139, which is hereby incorporated by reference.
  • diffuse injury I no visible pathology
  • diffuse injury II a. midline shift of 0 to 5 mm b. basal cisterns remain visible c. no high or mixed density lesions >25 cm 3
  • diffuse injury III swelling
  • midline shift of 0 to 5 mm b. basal cisterns compressed or completely effaced c. no high or mixed density lesions >25 cm 3
  • diffuse injury IV shift
  • the Marshall 234 score is a binary indication of whether or not an mTBI subject has a Marshall 2, 3, or 4 category score.
  • the Marshall 56 score is a binary indication of whether or not an mTBI subject has a Marshall 5 or 6 category score.
  • the first group for patients with a Marshall group of 1 is omitted to avoid issues of multicollinearity, since it is implicitly defined by the absence of the other two categories, e.g, when both binary variables for Marshall 234 and Marshall 56 are 0, it indicates that the mTBI subject belongs to Marshall group 1.
  • the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury.
  • the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two weeks of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two weeks of incurring the traumatic brain injury.
  • DB2/ 47566363.1 31 Attorney Ref. No.: 118547-5008-WO Referring to block 920, in some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury.
  • the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 days of incurring the traumatic brain injury
  • the radiographic modality finding of the mTBI subject is obtained within 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 days of incurring the traumatic brain injury.
  • an indication is provided, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit.
  • a mTBI subject incurs a chronic functional outcome deficit when they have a Glasgow Outcome Scale-Extended (GOS-E) score of less than or equal to 7, twelve months after incurring the initial traumatic brain injury.
  • GOS-E Glasgow Outcome Scale-Extended
  • a mTBI subject incurs a chronic functional outcome deficit when they have a GOSE score of less than or equal to 7, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 months after incurring the initial traumatic brain injury.
  • a mTBI subject incurs a chronic functional outcome deficit when they have a GOS-E score of less than or equal to 6, twelve months after incurring the initial traumatic brain injury.
  • a mTBI subject incurs a chronic functional outcome deficit when they have a GOS-E score of less than or equal to 6, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 months after incurring the initial traumatic brain injury.
  • GOS-E scores see McMahon, et al., 2014, “Symptomatology and Functional Outcome in Mild Traumatic Brain Injury: Results from the Prospective TRACK-TBI Study,” Journal of Neurotrauma 31, pp.26-33, which is hereby incorporated by reference.
  • an equivalent test is used to establish that a subject incurs a chronic functional outcome deficit.
  • GOS-E is a commonly used scale to assess the functional outcome of mTBI subjects following a traumatic brain injury.
  • GOS-E provides a structured way of categorizing mTBI DB2/ 47566363.1 32 Attorney Ref. No.: 118547-5008-WO subjects into different levels of recovery based on their level of disability and ability to function in daily life.
  • the GOS-E scores range from 1 to 8, with 1 representing death and 8 representing upper good recovery with the resumption of normal activities.
  • the following is a breakdown of Glasgow Outcome Scale-Extended scores: [00135] 1. Dead: The patient has died. [00136] 2. Vegetative State: The patient is unconscious and shows no evidence of awareness of self or environment. [00137] 3.
  • Lower Severe Disability The patient is conscious but requires constant nursing care due to severe neurological impairment. They are unable to live independently.
  • Upper Severe Disability The patient is conscious but requires assistance with daily living activities. They are unable to live independently but may be able to follow simple commands.
  • Lower Moderate Disability The patient is able to live independently but may require some assistance with certain tasks or have some cognitive or physical deficits.
  • Upper Moderate Disability The patient is able to live independently with minimal assistance but may have some persisting symptoms or disabilities that affect daily life.
  • Lower Good Recovery The patient has made a good recovery and is able to resume normal activities but may have some minor impairments or limitations.
  • the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit.
  • the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit.
  • the method further comprises obtaining a baseline heart rate, a diastolic blood pressure, a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject.
  • the method further comprises obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O 2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject.
  • the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O 2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject.
  • the providing the indication comprises: inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit.
  • the model is an Elastic Net model with a multinomial likelihood function.
  • the model is a nonnegative matrix factorization model.
  • the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof.
  • the inputting further comprises inputting one or more covariates into the model.
  • DB2/ 47566363.1 34 Attorney Ref. No.: 118547-5008-WO
  • the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof.
  • the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises treating the mTBI subject based on the indication.
  • the treating the mTBI subject based on the indication comprises administering psychotherapy to the mTBI subject.
  • Psychotherapy can be beneficial for individuals with mild traumatic brain injury (mTBI) to address cognitive, emotional, and behavioral challenges that may arise following the injury.
  • Cognitive behavioral therapy is one form of psychotherapy. CBT is a structured form of therapy that focuses on identifying and changing negative thought patterns and behaviors. It can help mTBI subjects manage symptoms such as anxiety, depression, and stress, as well as address cognitive difficulties such as memory problems and difficulty concentrating.
  • Psychoeducation is another form of psychotherapy. Psychoeducation involves providing information and education about mTBI, its symptoms, and its effects on cognitive and emotional functioning. This can help mTBI subjects better understand their condition and learn strategies for coping with and managing symptoms. See, for example, Caplain et al, 2019, “Efficacy of Psychoeducation and Cognitive Rehabilitation After Mild Traumatic Brain Injury for Preventing Post-concussional Syndrome in Individuals with High Risk of DB2/ 47566363.1 35 Attorney Ref.
  • Neuropsychological rehabilitation is another form of psychotherapy. Neuropsychological rehabilitation involves using cognitive training exercises and strategies to help improve cognitive functioning following mTBI. This may include exercises to improve memory, attention, and problem-solving skills, as well as compensatory strategies to help mTBI subjects work around areas of cognitive difficulty. See, for example, 2009, “Neuropsychological rehabilitation of mild traumatic brain injury,” Brain Injury 10:4, pp. 277-286, which is hereby incorporated by reference. [00158] Mindfulness-based interventions is another form of psychotherapy.
  • Mindfulness- based interventions such as mindfulness-based stress reduction (MBSR) or mindfulness- based cognitive therapy (MBCT) can help mTBI subjects reduce stress, improve mood, and enhance overall well-being by teaching them to cultivate present-moment awareness and acceptance. See, for example, Lovette et al., 2022, ‘“Hidden gains”? Measuring the impact of mindfulness-based interventions for people with mild traumatic brain injury: a scoping review,’ Brain Injury 36:9, pp.1059-1070, which is hereby incorporated by reference.
  • Supportive therapy is another form of psychotherapy. Supportive therapy involves providing emotional support and validation to mTBI subjects as they navigate the challenges of living with their injury.
  • the treating the mTBI subject based on the indication comprises administering physical therapy to the mTBI subject.
  • VRT Vestibular rehabilitation therapy
  • Many mTBI subjects experience vestibular dysfunction, which can cause symptoms such as dizziness, vertigo, and imbalance.
  • VRT is a specialized form of physical therapy designed to address these symptoms by using exercises and techniques to improve balance, coordination, and proprioception.
  • Oculomotor rehabilitation is another form of physical therapy that can be administered to mTBI subjects. Some mTBI subjects may experience visual disturbances and difficulties with eye movements following their injury. Oculomotor rehabilitation involves exercises and techniques to improve eye tracking, fixation, convergence, and other aspects of visual function. See, for example, Thiagarajan and Ciuffreda, 2013, “Effect of oculomotor rehabilitation on vergence responsivity in mild traumatic brain injury,” JRRD 50(9), pp.
  • Balance training is another form of physical therapy that can be administered to mTBI subjects.
  • Balance training exercises are commonly used in physical therapy to help mTBI subjects improve their balance and stability. These exercises may include standing on one leg, walking on uneven surfaces, and performing dynamic balance activities to challenge the vestibular and proprioceptive systems. See, for example, Kakade and Kanase, 2020, “Effect of Multidimensional Exercise Program for Improving Balance in Traumatic Brain Injury Patients,” Medico-legal Update, July-September 20(3), pp.155-161, which is hereby incorporated by reference.
  • Aerobic exercise is another form of physical therapy that can be administered to mTBI subjects.
  • Aerobic exercise such as walking, cycling, or swimming, can be beneficial for mTBI subjects to improve cardiovascular fitness, endurance, and overall physical health. Aerobic exercise has also been shown to have positive effects on mood and cognitive function. See, for example, Chin et al., 2015, “Improved Cognitive Performance Following Aerobic Exercise Training in People With Traumatic Brain Injury,” Archives of Physical Medicine and Rehabilitation 96(4), pp.754-759, which is hereby incorporated by reference. [00166]
  • Strengthening exercises are another form of physical therapy that can be administered to mTBI subjects. Strengthening exercises may be prescribed to mTBI subjects DB2/ 47566363.1 37 Attorney Ref. No.: 118547-5008-WO to help improve muscle strength, endurance, and functional capacity.
  • Gait training is another form of physical therapy that can be administered to mTBI subjects. Gait training involves working on walking patterns and mechanics to improve walking speed, stride length, and overall gait efficiency. This may include exercises to improve posture, balance, and coordination while walking.
  • Manual therapy is another form of physical therapy that can be administered to mTBI subjects.
  • Manual therapy techniques such as massage, joint mobilization, and soft tissue mobilization, may be used by physical therapists to help alleviate muscle tension, reduce pain, and improve mobility in mTBI subjects.
  • the treating the mTBI subject based on the indication comprises administering speech therapy to the mTBI subject.
  • Speech therapy for mTBI subjects focuses on addressing communication difficulties that may arise as a result of the injury.
  • One form of speech therapy for mTBI subjects is cognitive-communicative therapy.
  • Cognitive-communicative therapy targets cognitive processes involved in communication, such as attention, memory, executive function, and problem-solving.
  • This type of therapy aims to improve the ability to understand and use language effectively in various communication situations. See, for example, MacLennan, 2012, “Cognitive-communication rehabilitation for combat-related mild traumatic brain injury,” Journal of rehabilitation research and development 49(7):XI, which is hereby incorporated by reference.
  • Another form of speech therapy for mTBI subjects is language therapy.
  • Language therapy focuses on improving specific language skills that may be affected by mTBI, such as DB2/ 47566363.1 38 Attorney Ref. No.: 118547-5008-WO comprehension, expression, vocabulary, and grammar. Therapy may involve exercises to improve word finding, sentence structure, and overall language fluency.
  • speech therapy for mTBI subjects is speech sound production therapy. Some mTBI subjects may experience difficulties with speech sound production, such as articulation, phonation, and resonance. Speech sound production therapy involves exercises and techniques to improve speech clarity, accuracy, and intelligibility.
  • voice therapy may be recommended for mTBI subjects who experience changes in voice quality, pitch, volume, or resonance following their injury.
  • Treatment may include exercises to improve vocal fold function, breath support, and vocal resonance.
  • Fluency therapy is aimed at improving speech fluency and reducing stuttering or other disfluencies that may occur as a result of mTBI. Therapy may involve strategies to increase relaxation, slow speech rate, and enhance smoothness of speech production.
  • social communication skills training focuses on improving pragmatic language skills, such as turn-taking, topic maintenance, conversational repair, and nonverbal communication. This type of therapy may involve role-playing, video modeling, and structured social communication activities.
  • the treating the mTBI subject based on the indication comprises administering occupational therapy to the mTBI subject.
  • Occupational therapy for mTBI subjects focuses on improving functional abilities and facilitating a return to daily activities.
  • One form of occupational therapy for mTBI subjects is activities of daily living (ADL) Training.
  • ADL training medical practitioners such as occupational therapists work with mTBI subjects to improve their ability to perform basic self-care tasks, such as bathing, dressing, grooming, toileting, and feeding. This may involve breaking down tasks into smaller steps, using adaptive equipment, and practicing skills in a structured environment.
  • IADL instrumental activities of daily living
  • medical practitioners such as occupational therapists help mTBI subjects develop or regain skills needed for more complex daily activities, such as meal preparation, household chores, managing finances, and using transportation. This may include cognitive strategies, environmental modifications, and task- specific training.
  • Cognitive rehabilitation focuses on improving cognitive functions that may be affected by mTBI, such as attention, memory, executive function, and problem-solving. Medical practitioners such as occupational therapists may use various techniques, such as cognitive exercises, compensatory strategies, and environmental adaptations, to help mTBI patients manage cognitive difficulties and perform daily activities more independently.
  • mTBI mild traumatic brain injury
  • Another form of occupational therapy for mTBI subjects is sensory integration therapy. Some mTBI subjects may experience sensory processing difficulties, such as hypersensitivity or hyposensitivity to sensory input. Sensory integration therapy involves DB2/ 47566363.1 40 Attorney Ref. No.: 118547-5008-WO activities and exercises designed to help regulate sensory processing and improve tolerance to sensory stimuli, which can enhance participation in daily activities.
  • Another form of occupational therapy for mTBI subjects is fatigue management. Fatigue is a common symptom of mTBI that can significantly impact daily functioning.
  • Medical practitioners such as occupational therapists help mTBI subjects manage fatigue by teaching energy conservation techniques, scheduling rest breaks, prioritizing tasks, and modifying activities to reduce physical and cognitive exertion. See, for example, Xu et al.. 2017, “Complementary and alternative interventions for fatigue management after traumatic brain injury: a systematic review,” Therapeutic Advances in Neurological Disorders 10(5), pp.229-239, which is hereby incorporated by reference.
  • Another form of occupational therapy for mTBI subjects is community reintegration. Medical practitioners, such as occupational therapists support mTBI subjects in returning to meaningful activities and roles in their community, such as work, school, volunteering, and social activities.
  • the treating the mTBI subject based on the indication comprises administering melatonin or zolipidem to the mTBI subject. Insomnia is highly prevalent within the mTBI population and is a subtle, frequently persistent complaint that often goes undiagnosed.
  • the treating the mTBI subject comprises administering to the subject melatonin or zolipidem. See, for example, Zhou and Greenwald, 2018, “Update on Insomnia after Mild Traumatic Brain Injury,” Brain Sci.8(12), p.223, which is hereby incorporated by reference.
  • the treating the mTBI subject based on the indication comprises administering a selective serotonin reuptake inhibitor (SSRI) to the mTBI subject.
  • SSRI selective serotonin reuptake inhibitor
  • SSRIs are a class of antidepressant agents that inhibit the reuptake of serotonin by monoamine transporters in the presynaptic cell, allowing for increased availability of serotonin in the synaptic cleft and increased/repeated stimulation of serotonergic postsynaptic receptors, leading to increased synaptic signaling.
  • SSRIs increase DB2/ 47566363.1 41 Attorney Ref. No.: 118547-5008-WO extracellular serotonin (5-HT), activating the seven types of 5-HT receptors and many subtypes of each class, which can also be directly stimulated by the SSRI.
  • the treating the mTBI subject based on the indication comprises administering cyclobenzaprine to the mTBI subject.
  • Cyclobenzaprine is a muscle relaxant medication commonly prescribed for the relief of muscle spasms and associated pain. Cyclobenzaprine may be prescribed in certain situations for specific symptoms commonly experienced after mild traumatic brain injury, such as to treat muscle spasms, to provide pain relief, or to improve sleep.
  • Cyclobenzaprine can help reduce muscle spasms, which may alleviate associated discomfort and improve mobility.
  • Headaches and neck pain are common symptoms following mTBI, often due to muscle tension or strain. Cyclobenzaprine can help alleviate these symptoms by relaxing muscles and reducing associated pain.
  • Cyclobenzaprine has sedative effects, and it is sometimes prescribed to help individuals with mTBI who are experiencing sleep disturbances. Improving sleep quality can be important for overall recovery from mTBI, as adequate rest supports brain healing and cognitive function.
  • Cyclobenzaprine may be prescribed as part of an adjunctive therapy regimen alongside other medications or therapies to manage symptoms such as headaches, neck pain, and muscle stiffness associated with mTBI.
  • DB2/ 47566363.1 42 Attorney Ref. No.: 118547-5008-WO
  • the indication is that the mTBI subject will incur the chronic functional outcome deficit
  • the method further comprises adjusting a treatment regime of the mTBI subject based on the indication.
  • the adjusting the treatment regime comprises increasing or decreasing a frequency or duration of psychotherapy for the mTBI subject.
  • the adjusting the treatment regime comprises increasing or decreasing a frequency or a duration of physical therapy for the mTBI subject. Examples of physical therapy that may be adjusted in accordance with block 966 are described in block 948.
  • the adjusting the treatment regime comprises increasing or decreasing a frequency or a duration of speech therapy for the mTBI subject. Examples of speech therapy that may be adjusted in accordance with block 968 are described in block 950.
  • the adjusting the treatment regime comprises increasing or decreasing a frequency or a duration of occupational therapy for the mTBI subject. Examples of occupational therapy that may be adjusted in accordance with block 970 are described in block 952.
  • the adjusting the treatment regime comprises increasing or decreasing a dosage of melatonin or zolipidem given to the mTBI subject or increasing or decreasing a length of time the melatonin or zolipidem is given to the mTBI subject. Melatonin or zolipidem treatment regimens that may be adjusted in accordance with block 972 are described in block 954.
  • the adjusting the treatment regime comprises increasing or decreasing a dosage of a selective serotonin reuptake inhibitor given to the mTBI subject or increasing or decreasing a length of time the selective serotonin reuptake inhibitor is given to the mTBI subject.
  • Selective serotonin reuptake inhibitor treatment regimens that may be adjusted in accordance with block 974 are described in block 956.
  • the adjusting the treatment regime comprises increasing or decreasing a dosage of methylphenidate or amphetamine salt given to DB2/ 47566363.1 43 Attorney Ref.
  • the adjusting the treatment regime comprises increasing or decreasing a dosage of cyclobenzaprine give to the mTBI subject or increasing or decreasing a length of time the cyclobenzaprine is given to the mTBI subject.
  • Cyclobenzaprine treatment regimens that may be adjusted in accordance with block 978 are described in block 960.
  • Prognostic biomarkers could potentially advance precision therapeutics and provide insight into the causal underpinnings of chronic mTBI symptoms.
  • clusters of mTBI patients were characterized based on chronic psychiatric and functional symptom profiles 12 months after injury, and to predict cluster membership from baseline biological data.
  • This exampled aimed to advance the understanding and management of mTBI through two primary aims: [00204] First, to identify and characterize clusters of mTBI patients based on a variety of chronic symptoms prevalent at 12 months. This classification was achieved using unsupervised learning techniques to analyze data encompassing measures across psychiatric DB2/ 47566363.1 44 Attorney Ref. No.: 118547-5008-WO and daily living scales. By identifying these clusters, light was shed on the diverse trajectories of mTBI recovery, providing additional understanding of its long-term impact. [00205] Secondly, to build machine learning based models capable of predicting the 12- month outcome cluster for individual patients, using only baseline biological data.
  • PHQ-9 is a self-report, standardized depression rating scale is used for screening, diagnosing, and monitoring treatment response for major depressive disorder (MDD). See Kroenke et al., 2003, “The Patient Health Questionnaire-2: validity of a two-item depression screener,” Med Care 41(11), pp.1284-92, which is hereby incorporated by reference.
  • the PHQ-9 uses 9 items corresponding to the DSM-5 criteria for MDD and also assesses for DB2/ 47566363.1 45 Attorney Ref. No.: 118547-5008-WO psychosocial impairment.
  • Imputed scales were normalized using the min-max function before clustering.
  • NMF Nonnegative Matrix Factorization of chronic symptoms.
  • NMF Nonnegative Matrix Factorization
  • NMF provides a low-rank approximation of a non-negative matrix ⁇ ⁇ R ⁇ ⁇ ⁇ 0 ⁇ ⁇ , decomposing it into the product of two matrices: an ⁇ ⁇ ⁇ basis matrix ⁇ and a coefficient matrix ⁇ : ⁇ ⁇ ⁇ ⁇ [00213]
  • X is a ⁇ ⁇ ⁇ matrix of clinical symptom scores, where d is the number of items and n is the number of participants.
  • k is the rank of the approximation: the number of dimensions, and in this example the number of clusters.
  • W provides the weighting of each scale item in each of the k dimensions of the lower rank approximation
  • H provides the coordinates of each patient in the basis W.
  • Cluster membership is derived for each participant from the coefficients H by selecting the dimension with the highest H value.
  • DB2/ 47566363.1 46 Attorney Ref. No.: 118547-5008-WO [00214]
  • the NMF model was fit using an objective function derived from [46], which minimizes the Kullback-Leibler divergence between WH and X.
  • two sets of models ranging from 2 to 10 factors were estimated.
  • One set of models made use of the data X, while the other set of models was fit to data randomized across columns (scale items).
  • FIG. 2A A number of statistics measuring NMF fit for real and randomized data are plotted in Figure 2A. Each model was run 5 times, and the average of each fit statistic is shown for each rank. In particular, the silhouette score was of interest, as this is a measure of goodness of clustering [47]. In the case of NMF, silhouette scores can be generated for either basis or coefficient matrices. Because the goal is to cluster patients (as opposed to scale items), the coefficient silhouette was focused on.
  • the cophenetic chart plots consensus, while the dispersion, evar, residuals, and rss plot best ft.
  • lines with circles represent random data while lines with triangles represent real data.
  • line 202 is consensus type
  • line 204 is coefficients type
  • line 206 is coefficients type
  • line 208 is basis type
  • line 210 is consensus type
  • line 212 is basis type
  • line 214 is coefficients type
  • line 216 is coefficients type
  • line 218 is basis type
  • line 220 is basis type.
  • Figure 2B shows the difference in coefficient silhouette score between real versus random data clustering. For subsequent analyses, k was set to 5 clusters, as this produced the largest difference in silhouette score, and was at or near the highest difference for most other measures. [00215] It is worth noting that no claim that 5 is the true or optimal number of clusters for chronic mTBI symptoms is made in this example. The focus of this example is on predicting patterns in chronic symptoms.
  • the cluster labels were kept fixed in this example. The data indicate that 5 clusters is a reasonable place to start.
  • the fuzzy nature of NMF was used to estimate uncertainty about cluster membership. To do so, each patient’s NMF coordinates was normalized by their sum, so they summed to one. Treating these normalized coordinates as a multinomial probability distribution over clusters, the entropy for each patient was computed, which is then taken as a measure of cluster uncertainty.
  • the relationship between chronic symptom cluster and the following demographic and clinical variables was examined: age, gender, years of education, number of previous TBIs, and the severity of the worst previous TBI.
  • Computed tomography findings included Marshall score, contusion, subarachnoid hemorrhage, skull fracture, epidural hematoma, inraventricular hemorrhage, subdural hematoma, extra-axial hematoma, and edema.
  • Mean vitals, computed tomography scans, lab results, and blood biomarkers were collected immediately following hospitalization.
  • MRI data were collected 14.8 ⁇ 2.6 days (mean ⁇ SD) after injury, but these features were eventually excluded due to being missing in too many patients. See Figure 3A for the full list of measurements considered in this example. [00220] All 1377 patients with 12 month symptom data had some subset of baseline biological measurements.
  • Figure 3 shows the percent missing for each baseline measurement we considered, before (top) and after (bottom) measures were dropped and patients missing from more than 50% of observations. A more lenient cut-off was used for biological data compared to clinical scales due to the large number of missing observations overall.
  • Figure 3B shows the final list of features included after this cutoff. [00221] 1008 participants who met the cut-off for both chronic symptom and baseline biological data were left for use in the predictive models of this example. The remaining missing data were imputed, separately for training and test sets.
  • a nested process of fitting, hyperparameter tuning, and validating models was utilized to avoid data leakage and estimate uncertainty about generalization performance. Specifically, Monte Carlo simulation was performed, repeatedly taking random samples of train/test splits and using 10-fold Cross Validation (CV) to optimize hyperparameters within a sample training split, before evaluating on the test set for that sample.
  • the overall procedure for each sample was: [00225] Randomly sample train/test labels; [00226] Impute train and test sets separately with predictive mean matching; [00227] Optimize hyperparameters using 10-fold cross validation on training set; [00228] Evaluate AUC for each cluster on test set. [00229] This was repeated 100 times, providing a distribution of AUC scores and GLM coefficients for each cluster.
  • NMF coordinates as well as cluster labels were included in imputation, which was performed separately for training and test sets in a given sample. After imputation, data were transformed into a matrix, with factors being dummy coded [54]. Columns of the training and test data matrices were then mean-centered and variance normalized. A column of ones was added to each of the matrices to represent an intercept term. [00233] Model optimization [00234] For each Monte Carlo iteration, model hyperparameters were optimized with 10-fold cross validation using the training set alone and then fit to the training set using the optimized DB2/ 47566363.1 49 Attorney Ref. No.: 118547-5008-WO hyperparameters.
  • Elastic Net regression represents a class of generalized linear models solving the following optimization problem: m ⁇ i ⁇ n ⁇ ( ⁇ , ⁇ , ⁇ ) + ⁇ [( 1 ⁇ ⁇ ) ⁇ ⁇ ⁇ 2 2 /2 + ⁇ ⁇ ⁇ ⁇ 1 ]
  • ⁇ ( ⁇ , ⁇ , ⁇ ) is a negative log-likelihood loss function combining response variables ⁇ , model coefficients ⁇ , and a feature matrix ⁇ .
  • the form of this loss depends on the family of distributions assumed for the observations well as the link function (in this case, a multinomial likelihood with a softmax link).
  • both ⁇ and ⁇ are matrices, and ⁇ 2 and ⁇ 1 notation is used to refer to the Frobenius norm and the sum of ⁇ 1 norms across classes, respectively. Weights were provided for each class (1 ⁇ # ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ to handle class imbalance.
  • a 10-fold cross validation was used within the training set to find optimal hyperparameters for each Monte Carlo iteration.
  • a grid search was performed over values of ⁇ and ⁇ , fitting the optimal ⁇ coefficients for 10 cross validation folds while keeping the hyperparameters fixed to each ⁇ and ⁇ combination.
  • Hyperparameters were selected that led to the lowest held out deviance averaged across cross validation folds, and then the ⁇ parameters were refit to the whole training set with those hyperparameters. This process was repeated for each random train/test split.
  • Model evaluation [00238] For each train/test split, the test set performance of the model fit to the training set with cross validation-optimized hyperparameters was evaluated.
  • the heatmap illustrates that most psychiatric scale items were correlated with each other, and that there were clear patterns of increased correlations both within- and between- items. For example, multiple clusters within the PCL-5 were apparent along the diagonal, while GOS-E and MPAI items were correlated with each other. Satisfaction with life was negatively correlated with other scales.
  • Chronic symptom clustering [00244] NMF was used to cluster patients based on 12 month psychiatric symptoms. Results of fitting NMF with five dimensions are shown in Figure 5.
  • the basis matrix W provides the loadings of each dimension/cluster onto the original clinical scales.
  • Figure 5A shows a heatmap visualizing entries of W. The matrix is sparse, indicating that each factor is made up of a subset of scales.
  • Factor 1 was made up of PTSD symptoms from the PCL-5, especially from re-experiencing and avoidance clusters. The top five items for this factor were all from these two clusters, with the highest three ranked items being “Upset by reminders”; “Repeated memories”; and “Avoiding memories, thoughts, or feelings”.
  • Factor 2 was weighted most heavily for SWSL items, and thus could be considered an overall recovery cluster.
  • Factor 3 combined depression symptoms from PHQ-9, BSI-18, and PCL-5 scales.
  • Factor 4 reflected sleep symptoms and insomnia, combining items from PHQ-9, ISI, and PCL-5.
  • the primary goal of this example was to predict chronic symptom patterns from biological data collected at baseline after injury. Potential biomarker data collected from patients at baseline following injury were combined. This included vital signs, lab results, deriver computed tomography and MRI results, and three putative liquid biopsy-based biomarkers available on FITBIR: high sensitivity C-reactive protein (hs-CRP), S100 calcium binding protein B (S100B), and neuron-specific enolase (NSE).
  • hs-CRP high sensitivity C-reactive protein
  • S100B S100 calcium binding protein B
  • NSE neuron-specific enolase
  • Multinomial Elastic Net models were used to predict the 12-month symptom clusters from baseline biological data, performing a nested cross-validation loop to optimize hyperparameters, and Monte Carlo sampling of train/test splits to evaluate predictions out of sample.
  • PTSD cluster For the PTSD cluster (Cluster 1), the most predictive features were the presence of a Subarachnoid Hemorrhage, mean sodium levels, and neuron-specific enolase level. All features were non-zero in >90% of models. While neuron-specific enolase and sodium contributed positively to PTSD cluster DB2/ 47566363.1 53 Attorney Ref. No.: 118547-5008-WO assignment, subarachnoid hemorrhage contributed negatively. Presumably, this was to distinguish mTBI patients with chronic prominent PTSD symptoms from those with chronic functional or daily living problems. Unsurprisingly given their lack of predictive accuracy, no features were clearly associated with Depression, Sleep Impairment, or Life Satisfaction clusters (Figure 8).
  • Table 1 Results of logistic regression predicting PTSD cluster from demographic and biological variables.
  • the present invention can be implemented as a computer program product that includes a computer program mechanism embedded in a non-transitory computer-readable storage medium.
  • the computer program product could contain instructions for operating the user interfaces disclosed herein and described with respect to the Figures.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Neurology (AREA)
  • Physics & Mathematics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Physiology (AREA)
  • Artificial Intelligence (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Data Mining & Analysis (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Radiology & Medical Imaging (AREA)
  • Neurosurgery (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Evolutionary Computation (AREA)
  • Psychology (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

Methods for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit. One or more liquid biopsy -based baseline biomarker results of the mTBI subject are obtained. The one or more liquid biopsy -based baseline biomarker results comprises a baseline high-sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject. A radiographic modality finding of the mTBI subject is obtained. An indication is made, based at least on a combination of (i) the one or more liquid biopsy -based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit.

Description

Attorney Ref. No.: 118547-5008-WO SYSTEMS, METHODS, AND DEVICES FOR USING PROGNOSTIC INDICATORS OF CHRONIC OUTCOME IN TRAUMATIC BRAIN INJURY CROSS REFERENCE TO RELATED APPLICATIONS [0001] This application claims priority to United States Provisional Patent Application No. 63/450,325 entitled “SYSTEMS, METHODS, AND DEVICES FOR USING PROGNOSTIC INDICATORS OF CHRONIC FUNCTIONAL OUTCOME IN MILD TRAUMATIC BRAIN INJURY (TBI),” filed March 6, 2023, which is hereby incorporated by reference. This application also claims priority to United States Provisional Patent Application No. 63/472,084 entitled “SYSTEMS, METHODS, AND DEVICES FOR USING PROGNOSTIC INDICATORS OF CHRONIC FUNCTIONAL OUTCOME IN MILD TRAUMATIC BRAIN INJURY (TBI),” filed June 9, 2023, which is hereby incorporated by reference. TECHNICAL FIELD [0002] The present disclosure is directed to determining whether a mild traumatic brain injury (mTBI) subject will incur functional outcome deficit. BACKGROUND [0003] Mild Traumatic Brain Injury (mTBI) represents a significant public health challenge, accounting for approximately 3 million emergency room visits annually in the United States alone[1–4]. Patients diagnosed with mild Traumatic Brain Injury (mTBI) experience a wide variety of recovery trajectories and chronic symptoms. Despite its prevalence, the long-term outcomes following an mTBI are highly variable and unpredictable. A considerable proportion of patients, estimated between 20-50%, experience persistent or worsening functional impairments years after the injury [5–10]. Chronic psychiatric and functional problems are notably more common in patients with mTBI relative to those with non-head injuries [11–13], encompassing a broad spectrum of symptoms and diagnoses such as Post Traumatic Stress Disorder (PTSD) [14–16], cognitive deficits, insomnia, Major Depression (MDD) [17–20], and Generalized Anxiety (GAD) [21–23]. [0004] The underreporting of mTBIs suggests that these injuries could significantly contribute to persistent psychiatric symptoms in certain populations. For instance, studies in 1 DB2/ 47566363.1 Attorney Ref. No.: 118547-5008-WO veterans have highlighted mTBI as a strong predictor of chronic PTSD symptoms [24], underscoring the need to better understand the heterogeneity of chronic psychiatric symptoms following mTBI and the underlying biological factors. In particular, it is important to distinguish between different symptom trajectories across multiple psychiatric and functional measures, ideally collected soon after the injury to provide accurate and useful data. However, few studies have attempted to differentiate between cross-domain patterns of long- term outcomes in a predictive manner using baseline measurements from multiple biological modalities. Instead, most existing studies have focused on single modalities, individual symptom domains, and limited post-injury timeframes. For example, computed tomography measurements have been associated with 12-month functional outcomes [25], while other studies have variously identified statistical associations between biomarkers such as Glial fibrillary acidic protein (GFAP) and high-sensitivity C-reactive Protein (hsCRP) with either PTSD symptoms or functional outcome disabilities [26–28], typically at 6 months post- injury. [0005] Other literature emphasizes the complexity of predicting outcomes following mTBI and the importance of considering a wide range of factors. Studies such as those by Booker et al., 2019, “Predicting functional recovery after mild traumatic brain injury: the SHEFBIT cohort” Brain Injury 33:9, pp.1158-1165; Mikolić et al., 2020, “, Prediction of Global Functional Outcome and Post-Concussive Symptoms after Mild Traumatic Brain Injury: External Validation of Prognostic Models in the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study” Journal of Neurotrauma, Jan 2021, pp.196-209; and Mikolić et al., 2023, “Prognostic Models for Global Functional Outcome and Post-Concussion Symptoms Following Mild Traumatic Brain Injury: A Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study,” Journal of Neurotrama, August 2023, pp.1651-1670) have identified key prognostic indicators, including psychiatric history, initial Glasgow Coma Scale (GCS) scores, etiology of the injury, and early post-injury symptoms [29–31]. These findings highlight the need for developing and validating prognostic models that are specifically tailored to the mild TBI population, taking into account the diverse predictors of outcomes. [0006] Given the variability in measurements and outcomes across different reports, which often vary in their methods, the constellation of associations between mTBI, putative biomarkers, and chronic symptoms remains unclear. Therefore, a comprehensive, multi- DB2/ 47566363.1 2 Attorney Ref. No.: 118547-5008-WO modal approach that considers various biological markers, symptom domains, and timeframes post-injury is needed to fully understand the complex interplay between mTBI and chronic psychiatric symptoms. Such an approach will not only enhance understanding of mTBI but also inform the development of targeted interventions to improve long-term outcomes for individuals with mTBI. [0007] The ability to predict chronic symptoms with sufficient accuracy could significantly enhance individualized treatment planning, deepen our understanding of the etiology of chronic symptoms, and eventually aid in the development of diagnostic and monitoring biomarkers, as well as future treatment targets. In the context of TBI, whose symptoms cut across numerous diagnoses and domains, differentiating future symptom trajectories can help shed light on the myriad of psychiatric diagnoses and symptoms that commonly result from brain injury. SUMMARY [0008] A method for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit is provided. In some such embodiments, the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15. [0009] One or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a month of incurring the traumatic brain injury. The one or more liquid biopsy-based baseline biomarker results comprises a baseline high-sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject. In some embodiments, the one or more liquid biopsy-based baseline biomarker results comprises a neuron-specific enolase (NSE) measurement or a S100B measurement of the mTBI subject. [0010] A radiographic modality finding of the mTBI subject is obtained within a month of incurring the traumatic brain injury. [0011] In some embodiments, the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage. In some embodiments, the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. [0012] In some embodiments, the radiographic modality finding of the mTBI subject consists of (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. DB2/ 47566363.1 3 Attorney Ref. No.: 118547-5008-WO [0013] In some embodiments, the radiographic modality finding of the mTBI subject comprises a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture. [0014] In some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury. [0015] In some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury. [0016] An indication is provided, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit. [0017] In some embodiments, the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit. [0018] In some embodiments, the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit. [0019] In some embodiments, the method further comprises obtaining a baseline heart rate, a diastolic blood pressure, a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject. The indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline heart rate, the diastolic blood pressure, the respiratory rate, the temperature, the systolic blood pressure, or any combination thereof, of the mTBI subject. [0020] In some embodiments, the method further comprises obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject. The indication as to whether the mTBI subject will incur the DB2/ 47566363.1 4 Attorney Ref. No.: 118547-5008-WO chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject. [0021] In some embodiments, the providing the indication comprises: inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy- based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit. [0022] In some embodiments, the model is an Elastic Net model with a multinomial likelihood function. [0023] In some embodiments, the model is a nonnegative matrix factorization model. [0024] In some embodiments, the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof. [0025] In some embodiments, the inputting further comprises inputting one or more covariates into the model. [0026] In some embodiments, the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof. [0027] In some embodiments, the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises treating the mTBI subject based on the indication. [0028] In some embodiments, the treating the mTBI subject based on the indication comprises administering psychotherapy to the mTBI subject. In some embodiments, the DB2/ 47566363.1 5 Attorney Ref. No.: 118547-5008-WO treating the mTBI subject based on the indication comprises administering physical therapy to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering speech therapy to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering occupational therapy to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering melatonin or zolipidem to the mTBI subject. [0029] In some embodiments, the treating the mTBI subject based on the indication comprises administering a selective serotonin reuptake inhibitor to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering methylphenidate or amphetamine salt to the mTBI subject. In some embodiments, the treating the mTBI subject based on the indication comprises administering cyclobenzaprine to the mTBI subject. [0030] In some embodiments, the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises adjusting a treatment regime of the mTBI subject based on the indication. [0031] In some embodiments, the adjusting the treatment regime comprises increasing a frequency or duration of psychotherapy for the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a frequency or a duration of physical therapy for the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a frequency or a duration of speech therapy for the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a frequency or a duration of occupational therapy for the mTBI subject. [0032] In some embodiments, the adjusting the treatment regime comprises increasing a dosage of melatonin or zolipidem given to the mTBI subject or increasing a length of time the melatonin or zolipidem is given to the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a dosage of a selective serotonin reuptake inhibitor given to the mTBI subject or increasing a length of time the selective serotonin reuptake inhibitor is given to the mTBI subject. In some embodiments, the adjusting the treatment regime comprises increasing a dosage of methylphenidate or amphetamine salt given to the mTBI subject or increasing a length of time the methylphenidate or amphetamine salt is given to the mTBI subject In some embodiments, the adjusting the treatment regime comprises DB2/ 47566363.1 6 Attorney Ref. No.: 118547-5008-WO increasing a dosage of cyclobenzaprine give to the mTBI subject or increasing a length of time the cyclobenzaprine is given to the mTBI subject. [0033] The methods of the present disclosure have other features and advantages that will be apparent from, or are set forth in more detail in, the accompanying drawings, which are incorporated herein, and the following Detailed Description, which together serve to explain certain principles of exemplary embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS [0034] Figure 1 illustrates an exemplary system topology including a computer system, in accordance with an exemplary embodiment of the present disclosure for determining whether a mild traumatic brain injury (mTBI) subject will incur functional outcome deficit. [0035] Figures 2A and 2B illustrate selecting the number of Nonnegative Matrix Factorization (NMF) dimensions for 12 month symptom data. 2A) Goodness of fit statistics for NMF of real and randomized data for different numbers of dimensions, generated using the NMF R package [42] , in accordance with an embodiment of the present disclosure. 2B) The difference between the coefficient silhouette scores between real and randomized data for different dimensions, in accordance with an embodiment of the present disclosure. The coefficient silhouette score is a clusterability measure for the H coefficient matrix, in this case indicating patient coordinates. The data indicate that 5 clusters produce the largest difference. [0036] Figures 3A and 3B illustrate present and missing baseline biological measurements. 3A) Biological measures collected at the time of injury including vitals, labs, MRI, CT, and hsCRP, neuron-specific enolase (NSE), and S100B levels for all participants included in the symptom clustering. Missing data are represented with black cells. The value in parentheses is the percent of data missing for each measure, in accordance with an embodiment of the present disclosure. 3B) The same as above, after excluding participants and measurements missing more than 50% of observations, in accordance with an embodiment of the present disclosure. [0037] Figures 4A, 4B, and 4C illustrate a heatmap representing Pearson correlation coefficients between individual scale items for PHQ9, ISI, MPAI, SWLS, BSI, PCL5, and GOS-E assessments in 1377 mTBI patients, in accordance with an embodiment of the present DB2/ 47566363.1 7 Attorney Ref. No.: 118547-5008-WO disclosure. The labels for the horizontal axis share the same numbering as the labels for the vertical axis. [0038] Figures 5A, 5B, and 5C illustrate clustering of patients into chronic symptom sub- types. 5A) Heatmap illustrating weights from the NMF basis matrix, representing loadings for each scale item for each cluster. Clusters 1-5 corresponded to PTSD (re-experiencing and avoidance), Life Satisfaction, Depression, Sleep Disturbance, and Functional Outcome Deficits, respectively. 5B) tSNE dimensionality reduction of patient scores in the space of scale items. Shading corresponds to cluster assignments and opacity corresponds to assignment certainty, which was operationalized as the negative entropy of normalized NMF coordinates for each subject. 5C) Histogram showing the number of patients assigned to each cluster. [0039] Figure 6 illustrates the distribution of demographic and clinical variables across chronic symptom clusters in accordance with an embodiment of the present disclosure. All examined variables were significantly different across clusters (Fage(4, 1372) = 6.29, p < 0.0001; Fgender(4, 1372) = 2.93, p < 0.05; Fedu(4, 1352) = 14.7, p < 0.00000000001; Ftbi,n(4, 1372) = 8.11, p < 0.00001; Ftbi,sev.(4, 1372) = 9.13, p < 0.000001). [0040] Figures 7A, 7B, 7C, 7D, and 7E illustrate predicting 12 month symptom clusters with baseline biological data. 7A) Out-of-sample AUC scores for Elastic Net models trained on baseline data. Distributions represent uncertainty from 100 randomly divided train-test splits. Functional Outcome and PTSD clusters could be predicted from baseline biological data. 7B, 7C, 7D, and 7E) Feature importance for each measurement included in the model for clusters 1 (PTSD) and 5 (Functional Outcome), presented in terms of two statistics measured over random samples: the proportion of samples in which a coefficient was non-zero (7B and 7D), and the mean and standard error across samples (7C and 7E). The highest weighted features for the Functional Outcome cluster were hsCRP, subarachnoid hemorrhage, and intraventricular hemorrhage. The highest weighted features for the PTSD cluster were subarachnoid hemorrhage, sodium, and neuron-specific enolase. [0041] Figures 8A, 8B, 8C, 8D, 8E, 8F, 8G, 8H, 8I, and 8J illustrate feature importance for all cluster predictions (cluster 1: 8A and 8B; cluster 2: 8C and 8D; cluster 3: 8E and 8F; cluster 4: 8G and 8H; and cluster 5: 8I and 8J) in accordance with an embodiment of the present disclosure. Feature importance for each measurement included in the model, presented in terms of two statistics measured over random samples: the proportion of samples DB2/ 47566363.1 8 Attorney Ref. No.: 118547-5008-WO in which a coefficient was non-zero (left), and the mean and standard error across samples (right). Clusters 2 (Life Satisfaction), 3 (Depression), and 4 (Sleep Disturbance) could not be predicted from baseline data, and thus no features showed large average or a high proportion of non-zero GLM coefficients for these clusters. [0042] Figures 9A, 9B, 9C, 9D, 9E, 9F, and 9G describe determining whether a mild traumatic brain injury (mTBI) subject will incur functional outcome deficit in accordance with the present disclosure in which optional elements are indicated by dashed boxes. [0043] It should be understood that the appended drawings are not necessarily to scale, presenting a somewhat simplified representation of various features illustrative of the basic principles of the invention. DETAILED DESCRIPTION [0044] The present disclosure is directed to determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit [0045] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be apparent to one of ordinary skill in the art that the present disclosure may be practiced without these specific details. In other instances, well-known methods, procedures, and components have not been described in detail so as not to unnecessarily obscure aspects of the embodiments. [0046] It will also be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For instance, a first subject could be termed a second subject, and, similarly, a second subject could be termed a first subject, without departing from the scope of the present disclosure. The first subject and the second subject are both subjects, but they are not the same subject. [0047] The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the description of the invention and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term “and/or” as used herein refers to and DB2/ 47566363.1 9 Attorney Ref. No.: 118547-5008-WO encompasses any and all possible combinations of one or more of the associated listed items. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. [0048] The foregoing description included example systems, methods, techniques, instruction sequences, and computing machine program products that embody illustrative implementations. For purposes of explanation, numerous specific details are set forth in order to provide an understanding of various implementations of the inventive subject matter. It will be evident, however, to those skilled in the art that implementations of the inventive subject matter may be practiced without these specific details. In general, well-known instruction instances, protocols, structures, and techniques have not been shown in detail. [0049] The foregoing description, for purpose of explanation, has been described with reference to specific implementations. However, the illustrative discussions below are not intended to be exhaustive or to limit the implementations to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The implementations are chosen and described in order to best explain the principles and their practical applications, to thereby enable others skilled in the art to best utilize the implementations and various implementations with various modifications as are suited to the particular use contemplated. [0050] In the interest of clarity, not all of the routine features of the implementations described herein are shown and described. It will be appreciated that, in the development of any such actual implementation, numerous implementation-specific decisions are made in order to achieve the designer’s specific goals, such as compliance with use case- and business-related constraints, and that these specific goals will vary from one implementation to another and from one designer to another. Moreover, it will be appreciated that such a design effort might be complex and time-consuming, but nevertheless be a routine undertaking of engineering for those of ordering skill in the art having the benefit of the present disclosure. [0051] As used herein, the term “baseline” means soon after incurring a traumatic brain injury. In some embodiments, for example, a baseline measurement is taken of a mild traumatic brain injury subject in order to diagnose the subject’s injury. This baseline DB2/ 47566363.1 10 Attorney Ref. No.: 118547-5008-WO measurement can be, for instance, within a day, within a week, within two weeks or within a month of the subject incurring the brain injury. [0052] As used herein, the term “if” may be construed to mean “when” or “upon” or “in response to determining” or “in response to detecting,” depending on the context. Similarly, the phrase “if it is determined” or “if [a stated condition or event] is detected” may be construed to mean “upon determining” or “in response to determining” or “upon detecting [the stated condition or event]” or “in response to detecting [the stated condition or event],” depending on the context. [0053] As used herein, the term “about” or “approximately” can mean within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which can depend in part on how the value is measured or determined, e.g., the limitations of the measurement system. For example, “about” can mean within 1 or more than 1 standard deviation, per the practice in the art. “About” can mean a range of ± 20%, ± 10%, ± 5%, or ± 1% of a given value. Where particular values are described in the application and claims, unless otherwise stated, the term “about” means within an acceptable error range for the particular value. The term “about” can have the meaning as commonly understood by one of ordinary skill in the art. The term “about” can refer to ± 10%. The term “about” can refer to ± 5%. [0054] As used interchangeably herein, the term “classifier” or “model” refers to a machine learning model. [0055] In some embodiments, a model is an unsupervised learning model. One example of an unsupervised learning model is cluster analysis. In some embodiments, a model includes supervised machine learning. Nonlimiting examples of supervised learning models include, but are not limited to, logistic regression models, neural networks, support vector machines, Naive Bayes models, nearest neighbors models, random forest models, decision trees, boosted trees, multinomial logistic regression models, linear models, linear regression models, Gradient Boosting models, mixture models, hidden Markov models, Gaussian NB models, linear discriminant analysis models, or any combinations thereof. In some embodiments, a model is a multinomial classifier algorithm. In some embodiments, a model is a 2-stage stochastic gradient descent (SGD) model. In some embodiments, a model is a deep neural network (e.g., a deep-and-wide sample-level model). DB2/ 47566363.1 11 Attorney Ref. No.: 118547-5008-WO [0056] Neural networks. In some embodiments, the model is a neural network (e.g., a convolutional neural network and/or a residual neural network). Neural networks, also known as artificial neural networks (ANNs), include convolutional and/or residual neural networks (deep learning models). In some embodiments, neural networks are machine learning models that are trained to map an input dataset to an output dataset, where the neural network includes an interconnected group of nodes organized into multiple layers of nodes. For example, in some embodiments, the neural network architecture includes at least an input layer, one or more hidden layers, and an output layer. In some embodiments, the neural network includes any total number of layers, and any number of hidden layers, where the hidden layers function as trainable feature extractors that allow mapping of a set of input data to an output value or set of output values. In some embodiments, a deep learning model is a neural network including a plurality of hidden layers, e.g., two or more hidden layers. In some instances, each layer of the neural network includes a number of nodes (or “neurons”). In some embodiments, a node receives input that comes either directly from the input data or the output of nodes in previous layers, and performs a specific operation, e.g., a summation operation. In some embodiments, a connection from an input to a node is associated with a parameter (e.g., a weight and/or weighting factor). In some embodiments, the node sums up the products of all pairs of inputs, xi, and their associated parameters. In some embodiments, the weighted sum is offset with a bias, b. In some embodiments, the output of a node or neuron is gated using a threshold or activation function, f, which, in some instances, is a linear or non-linear function. In some embodiments, the activation function is, for example, a rectified linear unit (ReLU) activation function, a Leaky ReLU activation function, or other function such as a saturating hyperbolic tangent, identity, binary step, logistic, arcTan, softsign, parametric rectified linear unit, exponential linear unit, softPlus, bent identity, softExponential, Sinusoid, Sine, Gaussian, or sigmoid function, or any combination thereof. [0057] In some implementations, the weighting factors, bias values, and threshold values, or other computational parameters of the neural network, are “taught” or “learned” in a training phase using one or more sets of training data. For example, in some implementations, the parameters are trained using the input data from a training dataset and a gradient descent or backward propagation method so that the output value(s) that the ANN computes are consistent with the examples included in the training dataset. In some embodiments, the parameters are obtained from a back propagation neural network training process. DB2/ 47566363.1 12 Attorney Ref. No.: 118547-5008-WO [0058] Any of a variety of neural networks are suitable for use in accordance with the present disclosure. Examples include, but are not limited to, feedforward neural networks, radial basis function networks, recurrent neural networks, residual neural networks, convolutional neural networks, residual convolutional neural networks, and the like, or any combination thereof. In some embodiments, the machine learning makes use of a pre-trained and/or transfer-learned ANN or deep learning architecture. In some implementations, convolutional and/or residual neural networks are used, in accordance with the present disclosure. [0059] For instance, a deep neural network model includes an input layer, a plurality of individually parameterized (e.g., weighted) convolutional layers, and an output scorer. The parameters (e.g., weights) of each of the convolutional layers as well as the input layer contribute to the plurality of parameters (e.g., weights) associated with the deep neural network model. In some embodiments, at least 50 parameters, at least 100 parameters, at least 1000 parameters, at least 2000 parameters or at least 5000 parameters are associated with the deep neural network model. As such, deep neural network models require a computer to be used because they cannot be mentally solved. In other words, given an input to the model, the model output needs to be determined using a computer rather than mentally in such embodiments. See, for example, Krizhevsky et al., 2012, “Imagenet classification with deep convolutional neural networks,” in Advances in Neural Information Processing Systems 2, Pereira, Burges, Bottou, Weinberger, eds., pp.1097-1105, Curran Associates, Inc.; Zeiler, 2012 “ADADELTA: an adaptive learning rate method,” CoRR, vol. abs/1212.5701; and Rumelhart et al., 1988, “Neurocomputing: Foundations of research,” ch. Learning Representations by Back-propagating Errors, pp.696-699, Cambridge, MA, USA: MIT Press, each of which is hereby incorporated by reference. [0060] Neural networks, including convolutional neural networks, suitable for use as models are disclosed in, for example, Vincent et al., 2010, “Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion,” J Mach Learn Res 11, pp.3371-3408; Larochelle et al., 2009, “Exploring strategies for training deep neural networks,” J Mach Learn Res 10, pp.1-40; and Hassoun, 1995, Fundamentals of Artificial Neural Networks, Massachusetts Institute of Technology, each of which is hereby incorporated by reference. Additional example neural networks suitable for use as models are disclosed in Duda et al., 2001, Pattern Classification, Second Edition, John Wiley & Sons, Inc., New York; and Hastie et al., 2001, The Elements of Statistical Learning, Springer- Verlag, New York, each of which is hereby incorporated by reference in its entirety. DB2/ 47566363.1 13 Attorney Ref. No.: 118547-5008-WO Additional example neural networks suitable for use as models are also described in Draghici, 2003, Data Analysis Tools for DNA Microarrays, Chapman & Hall/CRC; and Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, each of which is hereby incorporated by reference in its entirety. [0061] Support vector machines. In some embodiments, the model is a support vector machine (SVM). SVMs suitable for use as models are described in, for example, Cristianini and Shawe-Taylor, 2000, “An Introduction to Support Vector Machines,” Cambridge University Press, Cambridge; Boser et al., 1992, “A training algorithm for optimal margin classifiers,” in Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory, ACM Press, Pittsburgh, Pa., pp.142-152; Vapnik, 1998, Statistical Learning Theory, Wiley, New York; Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; Duda, Pattern Classification, Second Edition, 2001, John Wiley & Sons, Inc., pp.259, 262-265; and Hastie, 2001, The Elements of Statistical Learning, Springer, New York; and Furey et al., 2000, Bioinformatics 16, 906- 914, each of which is hereby incorporated by reference in its entirety. When used for classification, SVMs separate a given set of binary labeled data with a hyper-plane that is maximally distant from the labeled data. For certain cases in which no linear separation is possible, SVMs work in combination with the technique of `kernels`, which automatically realizes a non-linear mapping to a feature space. The hyper-plane found by the SVM in feature space corresponds, in some instances, to a non-linear decision boundary in the input space. In some embodiments, the plurality of parameters (e.g., weights) associated with the SVM define the hyper-plane. In some embodiments, the hyper-plane is defined by at least 10, at least 20, at least 50, or at least 100 parameters and the SVM model requires a computer to calculate because it cannot be mentally solved. [0062] Naïve Bayes models. In some embodiments, the model is a Naive Bayes model. Naïve Bayes models suitable for use as models are disclosed, for example, in Ng et al., 2002, “On discriminative vs. generative classifiers: A comparison of logistic regression and naive Bayes,” Advances in Neural Information Processing Systems, 14, which is hereby incorporated by reference. A Naive Bayes model is any model in a family of “probabilistic models” based on applying Bayes’ theorem with strong (naïve) independence assumptions between the features. In some embodiments, they are coupled with Kernel density estimation. See, for example, Hastie et al., 2001, The elements of statistical learning: data DB2/ 47566363.1 14 Attorney Ref. No.: 118547-5008-WO mining, inference, and prediction, eds. Tibshirani and Friedman, Springer, New York, which is hereby incorporated by reference. [0063] Nearest neighbors. In some embodiments, a model is nearest neighbors. In some implementations, nearest neighbor models are memory-based and include no model to be fit. For nearest neighbors, given a query point x0 (a test subject), the k training points x(r), r, ... , k (here the training subjects) closest in distance to x0 are identified and then the point x0 is classified using the k nearest neighbors. In some embodiments, Euclidean distance in feature space is used to determine distance as ^^^^( ^^^^) = ‖ ^^^^( ^^^^) − ^^^^( ^^^^)‖. Typically, when the nearest neighbor model is used, the
Figure imgf000017_0001
the linear discriminant is standardized to have mean zero and variance 1. In some embodiments, the nearest neighbor rule is refined to address issues of unequal class priors, differential misclassification costs, and feature selection. Many of these refinements involve some form of weighted voting for the neighbors. For more information on nearest neighbor analysis, see Duda, Pattern Classification, Second Edition, 2001, John Wiley & Sons, Inc; and Hastie, 2001, The Elements of Statistical Learning, Springer, New York, each of which is hereby incorporated by reference. [0064] A k-nearest neighbor model is a non-parametric machine learning method in which the input consists of the k closest training examples in feature space. The output is a class membership. An object is classified by a plurality vote of its neighbors, with the object being assigned to the class most common among its k nearest neighbors (k is a positive integer, typically small). If k = 1, then the object is simply assigned to the class of that single nearest neighbor. See, Duda et al., 2001, Pattern Classification, Second Edition, John Wiley & Sons, which is hereby incorporated by reference. In some embodiments, the number of distance calculations needed to solve the k-nearest neighbor model is such that a computer is used to solve the model for a given input because it cannot be mentally performed. [0065] Random forest, decision tree, and boosted trees. In some embodiments, the model is a decision tree. Decision trees suitable for use as models are described generally by Duda, 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp.395-396, which is hereby incorporated by reference. Tree-based methods partition the feature space into a set of rectangles, and then fit a model (like a constant) in each one. In some embodiments, the decision tree is random forest regression. For example, one specific model is a classification and regression tree (CART). Other specific decision tree models include, but are not limited to, ID3, C4.5, MART, and Random Forests. CART, ID3, and C4.5 are described in Duda, DB2/ 47566363.1 15 Attorney Ref. No.: 118547-5008-WO 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp.396-408 and pp.411- 412, which is hereby incorporated by reference. CART, MART, and C4.5 are described in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, Chapter 9, which is hereby incorporated by reference in its entirety. Random Forests are described in Breiman, 1999, “Random Forests--Random Features,” Technical Report 567, Statistics Department, U.C. Berkeley, September 1999, which is hereby incorporated by reference in its entirety. In some embodiments, the decision tree model includes at least 10, at least 20, at least 50, or at least 100 parameters (e.g., weights and/or decisions) and requires a computer to calculate because it cannot be mentally solved. [0066] Regression. In some embodiments, the model uses a regression formula. In some embodiments, the regression formula is determined using any type of regression. For example, in some embodiments, the regression is logistic regression. In some embodiments, the regression is logistic regression with lasso, L2 or elastic net regularization. In some embodiments, those extracted features that have a corresponding regression coefficient that fails to satisfy a threshold value are pruned (removed from) consideration. In some embodiments, a generalization of the logistic regression model that handles multicategory responses is used as the model. Logistic regression is disclosed in Agresti, An Introduction to Categorical Data Analysis, 1996, Chapter 5, pp.103-144, John Wiley & Son, New York, which is hereby incorporated by reference. In some embodiments, the model makes use of a regression model disclosed in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York. In some embodiments, the logistic regression model includes at least 10, at least 20, at least 50, at least 100, or at least 1000 parameters (e.g., weights) and requires a computer to calculate because it cannot be mentally solved. [0067] Linear discriminant analysis models. In some embodiments, linear discriminant analysis (LDA), normal discriminant analysis (NDA), or discriminant function analysis is a generalization of Fisher’s linear discriminant, a method used in statistics, pattern recognition, and machine learning to find a linear combination of features that characterizes or separates two or more classes of objects or events. In some embodiments, the resulting combination is used as the model (linear model) in some embodiments of the present disclosure. [0068] Mixture model and Hidden Markov model. In some embodiments, the model is a mixture model, such as that described in McLachlan et al., Bioinformatics 18(3):413-422, 2002. In some embodiments, in particular, those embodiments including a temporal DB2/ 47566363.1 16 Attorney Ref. No.: 118547-5008-WO component, the model is a hidden Markov model such as described by Schliep et al., 2003, Bioinformatics 19(1):i255-i263. [0069] Clustering. In some embodiments, the model is an unsupervised clustering model. In some embodiments, the model is a supervised clustering model. Clustering algorithms suitable for use as models are described, for example, at pages 211-256 of Duda and Hart, Pattern Classification and Scene Analysis, 1973, John Wiley & Sons, Inc., New York, (hereinafter “Duda 1973”) which is hereby incorporated by reference in its entirety. s an illustrative example, in some embodiments, the clustering problem is described as one of finding natural groupings in a dataset. To identify natural groupings, two issues are addressed. First, a way to measure similarity (or dissimilarity) between two samples is determined. This metric (e.g., similarity measure) is used to ensure that the samples in one cluster are more like one another than they are to samples in other clusters. Second, a mechanism for partitioning the data into clusters using the similarity measure is determined. One way to begin a clustering investigation is to define a distance function and to compute the matrix of distances between all pairs of samples in the training set. If distance is a good measure of similarity, then the distance between reference entities in the same cluster is significantly less than the distance between the reference entities in different clusters. However, in some implementations, clustering does not use a distance metric. For example, in some embodiments, a nonmetric similarity function s(x, x') is used to compare two vectors x and x'. In some such embodiments, s(x, x') is a symmetric function whose value is large when x and x' are somehow “similar.” Once a method for measuring “similarity” or “dissimilarity” between points in a dataset has been selected, clustering uses a criterion function that measures the clustering quality of any partition of the data. Partitions of the dataset that extremize the criterion function are used to cluster the data. Particular exemplary clustering techniques contemplated for use in the present disclosure include, but are not limited to, hierarchical clustering (agglomerative clustering using a nearest-neighbor algorithm, farthest-neighbor algorithm, the average linkage algorithm, the centroid algorithm, or the sum-of-squares algorithm), k-means clustering, fuzzy k-means clustering algorithm, and Jarvis-Patrick clustering. In some embodiments, the clustering includes unsupervised clustering (e.g., with no preconceived number of clusters and/or no predetermination of cluster assignments). [0070] Ensembles of models and boosting. In some embodiments, an ensemble (two or more) of models is used. In some embodiments, a boosting technique such as AdaBoost is DB2/ 47566363.1 17 Attorney Ref. No.: 118547-5008-WO used in conjunction with many other types of learning algorithms to improve the performance of the model. In this approach, the output of any of the models disclosed herein, or their equivalents, is combined into a weighted sum that represents the final output of the boosted model. In some embodiments, the plurality of outputs from the models is combined using any measure of central tendency known in the art, including but not limited to a mean, median, mode, a weighted mean, weighted median, weighted mode, etc. In some embodiments, the plurality of outputs is combined using a voting method. In some embodiments, a respective model in the ensemble of models is weighted or unweighted. [0071] As used herein, the term “parameter” refers to any coefficient or, similarly, any value of an internal or external element (e.g., a weight and/or a hyperparameter) in an algorithm, model, regressor, and/or classifier that can affect (e.g., modify, tailor, and/or adjust) one or more inputs, outputs, and/or functions in the algorithm, model, regressor and/or classifier. For example, in some embodiments, a parameter refers to any coefficient, weight, and/or hyperparameter that can be used to control, modify, tailor, and/or adjust the behavior, learning, and/or performance of an algorithm, model, regressor, and/or classifier. In some instances, a parameter is used to increase or decrease the influence of an input (e.g., a feature) to an algorithm, model, regressor, and/or classifier. As a nonlimiting example, in some embodiments, a parameter is used to increase or decrease the influence of a node (e.g., of a neural network), where the node includes one or more activation functions. Assignment of parameters to specific inputs, outputs, and/or functions is not limited to any one paradigm for a given algorithm, model, regressor, and/or classifier but can be used in any suitable algorithm, model, regressor, and/or classifier architecture for a desired performance. In some embodiments, a parameter has a fixed value. In some embodiments, a value of a parameter is manually and/or automatically adjustable. In some embodiments, a value of a parameter is modified by a validation and/or training process for an algorithm, model, regressor, and/or classifier (e.g., by error minimization and/or backpropagation methods). In some embodiments, an algorithm, model, regressor, and/or classifier of the present disclosure includes a plurality of parameters. In some embodiments, the plurality of parameters is n parameters, where: n ≥ 2; n ≥ 5; n ≥ 10; n ≥ 25; n ≥ 40; n ≥ 50; n ≥ 75; n ≥ 100; n ≥ 125; n ≥ 150; n ≥ 200; n ≥ 225; n ≥ 250; n ≥ 350; n ≥ 500; n ≥ 600; n ≥ 750; n ≥ 1,000; n ≥ 2,000; n ≥ 4,000; n ≥ 5,000; n ≥ 7,500; n ≥ 10,000; n ≥ 20,000; n ≥ 40,000; n ≥ 75,000; n ≥ 100,000; n ≥ 200,000; n ≥ 500,000, n ≥ 1 x 106, n ≥ 5 x 106, or n ≥ 1 x 107. As such, some embodiments of the algorithms, models, regressors, and/or classifiers of the present disclosure cannot be DB2/ 47566363.1 18 Attorney Ref. No.: 118547-5008-WO mentally performed. In some embodiments n is between 10,000 and 1 x 107, between 100,000 and 5 x 106, or between 500,000 and 1 x 106. In some embodiments, the algorithms, models, regressors, and/or classifier of the present disclosure operate in a k-dimensional space, where k is a positive integer of 5 or greater (e.g., 5, 6, 7, 8, 9, 10, etc.). As such, some embodiments of the algorithms, models, regressors, and/or classifiers of the present disclosure cannot be mentally performed. [0072] Distributed Computer System. In the present disclosure, unless expressly stated otherwise, descriptions of devices and systems will include implementations of one or more computers. For instance, and for purposes of illustration in Figure 1, a computer system 100 is represented as single device that includes all the functionality of the computer system 100, such as one or more processing units (e.g., one or more central processing units and/or one or more graphics processing units). However, the present disclosure is not limited thereto. For instance, in some embodiments, the functionality of the computer system 100 is spread across any number of networked computers and/or reside on each of several networked computers and/or by hosted on one or more virtual machines and/or containers at a remote location accessible across a communications network (e.g., communications network 106 of Figure 1). One of skill in the art will appreciate that a wide array of different computer topologies is possible for the computer system 100, and other devices and systems of the preset disclosure, and that all such topologies are within the scope of the present disclosure. Moreover, rather than relying on a physical communications network 106, the illustrated devices and systems may wirelessly transmit information between each other. As such, the exemplary topology shown in Figure 1 merely serves to describe the features of an embodiment of the present disclosure in a manner that will be readily understood to one of skill in the art. [0073] Figure 1 depicts a block diagram of a distributed computer system (e.g., computer system 100) according to some embodiments of the present disclosure. The computer system 100 at least facilitates communicating one or more instructions for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit. [0074] In some embodiments, the communication network 106 optionally includes the Internet, one or more local area networks (LANs), one or more wide area networks (WANs), other types of networks, or a combination of such networks. [0075] Examples of communication networks 106 include the World Wide Web (WWW), an intranet and/or a wireless network, such as a cellular telephone network, a wireless local area DB2/ 47566363.1 19 Attorney Ref. No.: 118547-5008-WO network (LAN) and/or a metropolitan area network (MAN), and other devices by wireless communication. The wireless communication optionally uses any of a plurality of communications standards, protocols and technologies, including Global System for Mobile Communications (GSM), Enhanced Data GSM Environment (EDGE), high-speed downlink packet access (HSDPA), high-speed uplink packet access (HSUPA), Evolution, Data-Only (EV-DO), HSPA, HSPA+, Dual-Cell HSPA (DC-HSPDA), long term evolution (LTE), near field communication (NFC), wideband code division multiple access (W-CDMA), code division multiple access (CDMA), time division multiple access (TDMA), Bluetooth, Wireless Fidelity (Wi-Fi) (e.g., IEEE 802.11a, IEEE 802.11ac, IEEE 802.11ax, IEEE 802.11b, IEEE 802.11g and/or IEEE 802.11n), voice over Internet Protocol (VoIP), Wi- MAX, a protocol for e-mail (e.g., Internet message access protocol (IMAP) and/or post office protocol (POP)), instant messaging (e.g., extensible messaging and presence protocol (XMPP), Session Initiation Protocol for Instant Messaging and Presence Leveraging Extensions (SIMPLE), Instant Messaging and Presence Service (IMPS)), and/or Short Message Service (SMS), or any other suitable communication protocol, including communication protocols not yet developed as of the filing date of this document. [0076] In various embodiments, the computer system 100 includes one or more processing units (CPUs), such as one or more central processing units (CPUs) and/or one or more graphic processing units (GPUs) 102 a network or other communications interface 104, and a memory 112. [0077] In some embodiments, the computer system 100 includes a user interface 106. The user interface 106 typically includes a display 108 for presenting media. In some embodiments, the display 108 is integrated within the computer systems (e.g., housed in the same chassis as the CPU 102 and the memory 112). In some embodiments, the computer system 100 includes one or more input device(s) 110 that allow a subject to interact with the computer system 100. In some embodiments, the one or more input devices 110 include a keyboard, a mouse, and/or other input mechanisms. Alternatively, or in addition, in some embodiments, the display 108 includes a touch-sensitive surface (e.g., where display 108 is a touch-sensitive display or computer system 100 includes a touch pad). [0078] In some embodiments, the computer system 100 presents media to a user through the display 108. Examples of media presented by the display 108 include one or more images, a video, audio (e.g., waveforms of an audio sample), or a combination thereof. In typical embodiments, the one or more images, the video, the audio, or the combination thereof is DB2/ 47566363.1 20 Attorney Ref. No.: 118547-5008-WO presented by the display 108 through a client application (e.g., a client application hosted by the computer system 100, a client application accessible through the computer system 100, etc.). In some embodiments, the audio is presented through an external device (e.g., speakers, headphones, input/output (I/O) subsystem, etc.) that receives audio information from the computer system 100 and presents audio data based on this audio information. In some embodiments, the user interface 106 also includes an audio output device, such as speakers or an audio output for connecting with speakers, earphones, or headphones. [0079] The memory 112 includes high-speed random access memory, such as DRAM, SRAM, DDR RAM, or other random access solid state memory devices, and optionally also includes non-volatile memory, such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid state storage devices. The memory 112 may optionally include one or more storage devices remotely located from the CPU(s) 102. The memory 112, or alternatively the non-volatile memory device(s) within memory 112, includes a non-transitory computer readable storage medium. Access to memory 112 by other components of the computer system 100, such as the CPU(s) 102, is, optionally, controlled by a controller. In some embodiments, the memory 112 include mass storage that is remotely located with respect to the CPU(s) 102. In other words, some data stored in the memory 112 may in fact be hosted on devices that are external to the computer system 100, but that can be electronically accessed by the computer system 100 over an Internet, intranet, or other form of network 106 or electronic cable using communication interface 104. [0080] In some embodiments, the memory 112 of the computer system 100 stores: • an operating system 120 (e.g., ANDROID, iOS, DARWIN, RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system such as VxWorks) that includes procedures for handling various basic system services; • an electronic address associated with the computer system 100 that identifies the computer system 100 (e.g., within the communication network 106); • a control module 122 including one or more models 124 for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit; and • optionally, a client application for presenting information (e.g., media) using a display 108 of the computer system 100. DB2/ 47566363.1 21 Attorney Ref. No.: 118547-5008-WO [0081] In some embodiments, the control module 122 includes one or more models 124 that are configured to perform one or more elements of a method of the present disclosure. [0082] Each of the above identified modules and applications correspond to a set of executable instructions for performing one or more functions described above and the methods described in the present disclosure (e.g., the computer-implemented methods and other information processing methods described herein). These modules (e.g., sets of instructions) need not be implemented as separate software programs, procedures or modules, and thus various subsets of these modules are, optionally, combined or otherwise re-arranged in various embodiments of the present disclosure. In some embodiments, the memory 112 optionally stores a subset of the modules and data structures identified above. Furthermore, in some embodiments, the memory 112 stores additional modules and data structures not described above. [0083] It should be appreciated that the computer system of Figure 1 is only one example of a computer system 100, and that the computer system 100 optionally has more or fewer components than shown, optionally combines two or more components, or optionally has a different configuration or arrangement of the components. The various components shown in Figure 1 are implemented in hardware, software, firmware, or a combination thereof, including one or more signal processing and/or application specific integrated circuits. [0084] Now that a general topology of a computer system 100 has been described in accordance with various embodiments of the present disclosures, details regarding some processes and methods of the present disclosure will be described in conjunction with Figure 11. [0085] Referring to block 900, a method for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit is provided. In some embodiments, TBI is any disruption in consciousness, motor function, sensory function, autonomic function, or ordinary brain function, whether transient or permanent, that is the result of a traumatic injury to the head. In some embodiments, the severity of TBI is graded using the Glasgow Coma Scale (GCS), where mild TBI is defined as GCS > 13, moderate TBI as GCS 9–12, and severe TBI as GCS 3–8. See Teasdale and Jennett, 1974, “Assessment of coma and impaired consciousness: A practical scale,” Lancet 304, 81–84, which is hereby incorporated by reference. DB2/ 47566363.1 22 Attorney Ref. No.: 118547-5008-WO [0086] Referring to block 902, in some embodiments, the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15. For information on the GCS score, see Teasdale et al., 2014, “The glasgow coma scale at 40 years: Standing the test of time,” The Lancet Neurology 13, pp.844-854 which is hereby incorporated by reference. The GCS is a neurological scale used to assess the level of consciousness of patients. The scale evaluates three aspects of neurological function: eye response, verbal response, and motor response. Each aspect is assessed independently and then combined to give a total score. The total score ranges from 3 to 15, with 3 being the worst possible score (deep coma or brain death) and 15 being the best possible score (fully awake and oriented). The GCS is commonly used in emergency and critical care settings to assess the severity of brain injury and monitor changes in neurological status over time. It helps healthcare providers make decisions regarding treatment and prognosis for patients with traumatic brain injury, stroke, and other conditions affecting consciousness. [0087] In some embodiments, the mTBI subject has a baseline GCS score in the range of 13- 15 within the month of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15 within a week of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 13-15 within two days of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 13-15 at the time of incurring a traumatic brain injury. [0088] In some embodiments, the mTBI subject has a baseline GCS score in the range of 12- 15 within the month of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 12-15 within a week of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 12-15 within two days of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 12-15 at the time of incurring a traumatic brain injury. [0089] In some embodiments, the mTBI subject has a baseline GCS score in the range of 11- 15 within the month of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 11-15 within a week of incurring a traumatic brain injury. In some embodiments, the mTBI subject has a baseline GCS score in the range of 11-15 within two days of incurring a traumatic brain injury. In some embodiments, the DB2/ 47566363.1 23 Attorney Ref. No.: 118547-5008-WO mTBI subject has a baseline GCS score in the range of 11-15 at the time of incurring a traumatic brain injury. [0090] Referring to block 904, one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a month of incurring the traumatic brain injury. The one or more liquid biopsy-based baseline biomarker results comprises a baseline high- sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject. [0091] hs-CRP is a marker of inflammation in the body and, in some embodiments, is measured through a blood test. For instance, in some embodiments, to obtain the hs-CRP measurement of the mTBI subject, a healthcare provider draws a sample of blood from a vein, usually in the arm, using a needle and syringe. The procedure is similar to other blood tests. The concentration of hs-CRP is then measured in the blood sample in e.g., in units of milligrams per liter (mg/L) of blood. This measurement reflects the amount of hs-CRP present in the bloodstream at the time the sample was collected. See Xu et al., 2021, “High- Sensitivity C-Reactive Protein Is a Prognostic Biomarker of Six-Month Disability after Traumatic Brain Injury: Results from the TRACK-TBI Study,” Journal of Neurotrauma 38, pp.918-927, which is hereby incorporated by reference. [0092] Referring to block 906, in some embodiments, the one or more liquid biopsy-based baseline biomarker results comprises a neuron-specific enolase (NSE) measurement or a S100B measurement of the mTBI subject. [0093] NSE is an enzyme that is primarily found in neurons, particularly in the central nervous system (brain and spinal cord). It is also present in neuroendocrine cells and certain peripheral nerves. NSE plays a role in glycolysis, the metabolic process that breaks down glucose to produce energy within cells. NSE is used as a biomarker for various neurological conditions, particularly those involving neuronal damage or death. Elevated levels of NSE in the blood or cerebrospinal fluid (CSF) can indicate neuronal injury or damage. Therefore, NSE measurement is often used in the diagnosis, prognosis, and monitoring of neurological disorders such as TBI. Measurement of NSE typically involves a blood test or analysis of cerebrospinal fluid. [0094] S100B is a protein belonging to the S100 protein family, which consists of calcium- binding proteins found predominantly in cells of the nervous system and some other tissues. S100B, specifically, is primarily expressed by astrocytes, a type of glial cell in the brain, although it is also present in other cells such as melanocytes. S100B is utilized as a DB2/ 47566363.1 24 Attorney Ref. No.: 118547-5008-WO biomarker for traumatic brain injury. Similar to NSE, elevated levels of S100B in the blood or CSF can indicate neuronal damage or injury. Measurement of S100B is used in the diagnosis of TBI. Similar to NSE, measurement of S100B typically involves a blood test or analysis of cerebrospinal fluid. [0095] In some embodiments, the one or more liquid biopsy-based baseline biomarker results are obtained from one or more liquid biopsy samples acquired from the mTBI subject. In some embodiments the one or more liquid biopsy samples are from the blood, whole blood, plasma, serum, urine, cerebrospinal fluid, fecal material, saliva, sweat, tears, pleural fluid, pericardial fluid, and/or the peritoneal fluid of the mTBI subject [0096] In some embodiments, a liquid biopsy sample in the one or more liquid biopsy samples has a volume of between 100 μL and 950 μL. For example, in some embodiments, the liquid biopsy sample has a volume of between 100 μL and 950 μL, between 120 μL and 900 μL, between 140 μL and 800 μL, between 160 μL and 700 μL, between 180 μL and 600 μL, between 200 μL and 550 μL ¸or between 220 μL and 400 μL. [0097] In some embodiments, the liquid biopsy sample in the one or more liquid biopsy samples has a volume of between 10 μL and 300 μL. For example, in some embodiments, the liquid biopsy sample has a volume of between 15 μL and 275 μL, between 20 μL and 250 μL, between 25 μL and 225 μL, between 30 μL and 250 μL, between 35 μL and 225 μL, between 40 μL and 200 μL ¸or between 50 μL and 125 μL. [0098] In some embodiments, the liquid biopsy sample in the one or more liquid biopsy samples has a volume of from about 1 mL to about 50 mL. For example, in some embodiments, the liquid biopsy sample has a volume of about 1 mL, about 2 mL, about 3 mL, about 4 mL, about 5 mL, about 6 mL, about 7 mL, about 8 mL, about 9 mL, about 10 mL, about 11 mL, about 12 mL, about 13 mL, about 14 mL, about 15 mL, about 16 mL, about 17 mL, about 18 mL, about 19 mL, about 20 mL, or greater. [0099] Referring to block 908, a radiographic modality finding of the mTBI subject is obtained within a month of incurring the traumatic brain injury. [00100] In some embodiments, magnetic resonance imagining (MRI) is used to obtain the radiographic modality finding. In some embodiments this MRI involves subjecting the mTBI subject to a main magnetic field (B0) (e.g., 1.5 T or 3.0 T), gradient coils providing small magnetic field variation for spatial localization, and radiofrequency proton excitation with DB2/ 47566363.1 25 Attorney Ref. No.: 118547-5008-WO subsequent decay, signal detection, and postprocessing. By varying the frequency and timing of the magnetic pulses, it is possible to localize the changes in radiofrequency signals to a particular place within the body. Additionally, the rate at which different protons return to their resting state helps distinguish separate tissues. Postprocessing of these signals allows for the construction of images that can be used to obtain the radiographic modality finding. See Nadel, 2021, “Emerging Utility of Applied Magnetic Resonance Imaging in the Management of Traumatic Brain Injury,” Med. Sci.9, 10, 0. https://doi.org/10.3390/medsci9010010, which is hereby incorporated by reference. [00101] In some embodiments, functional MRI (fMRI) is used to obtain the radiographic modality finding. fMRI is an application of MR technology whereby clinicians can differentially identify areas of brain activation during specific tasks or in the resting state. The basis of fMRI relies on a blood oxygenation level–dependent effect. See, Buxton, 2009, Introduction to Functional Magnetic Resonance Imaging: Principles and Techniques, Cambridge University Press: Cambridge, UK, which is hereby incorporated by reference. This effect comprises two primary assumptions: (1) as oxygenated hemoglobin transitions to deoxyhemoglobin or vice versa, there is a small but detectable change in the magnetic properties of the heme related to iron oxidation status; and (2) increased neuronal activation in a particular brain region has an associated increase in local cerebral blood flow and oxygen extraction. Taken together, as certain regions of the brain are increasingly activated during a task, there is a local increase in cerebral blood flow and oxygenated hemoglobin transitions to deoxyhemoglobin at increased rates. This produced a subtle change in the magnetic signals in that region, and signal averages over long imaging time can be detected on MRI. See, Logothetis et al., 2001, “Neurophysiological investigation of the basis of the fMRI signal,” Nature 412, pp.150–157, which is hereby incorporated by reference. [00102] In some embodiments, diffusion tensor imaging (DTI) is used to obtain the radiographic modality finding. As with DWI, in the acquisition of DTI images, the diffusion of water molecules is quantified within the tissue slab. However, specifically with DTI, multiple parameters are acquired, including the rate at which water molecules diffuse in the tissues as well as the direction of that diffusion. These parameters are acquired for each voxel of the MR image. From those data, specific measures are calculated to describe water diffusion in tissues, including anisotropy and diffusivity. See, for example, Smith et al., 2019, “Advanced neuroimaging in traumatic brain injury: An overview,” Neurosurg. Focus 47, E17; Alexander et al., 2007, “Diffusion tensor imaging of the brain, DB2/ 47566363.1 26 Attorney Ref. No.: 118547-5008-WO Neurotherapeutics 4, pp; 316–329 and Le Bihan et al., 2001, “Diffusion tensor imaging: Concepts and applications,” J. Magn. Reson. Imaging 13, pp.534–546, each of which is hereby incorporated by reference. Without intending to be limited by any particular theory, these values are thought to correlate with the biological integrity of the brain’s white matter, as water will more readily diffuse down intact tracts. Higher anisotropy and lower diffusivity are correlated with greater white matter integrity. For this reason, these values have clinical implications. Additionally, because the data collected include a directional component to the diffusion, postprocessing allows for the identification and tracing of specific axonal tracts within brain tissue. [00103] In some embodiments, magnetic resonance perfusion (MRP) is used to obtain the radiographic modality finding. MRP is a magnetic resonance technique that is used to determine and track intracerebral blood flow dynamics. Dynamic susceptibility contrast (DSC) imaging is a perfusion technique in which gadolinium contrast is administered and the decrease in T2 * signal (susceptibility) is quantified as the contrast bolus passes through the brain. See, Petrella and Provenzale, 2000, “MR Perfusion Imaging of the Brain: Techniques and applications,” Am. J. Roentgenol.175, pp.207–219, which is hereby incorporated by reference. Calculated parameters from DSC include cerebral blood volume, cerebral blood flow, mean transit time, and time to peak of the contrast bolus through the tissues. A similar perfusion technique, known as dynamic contrast-enhanced (DCE) imaging, relies on the T1- shortening effects of the gadolinium and, as such, signal increases as the bolus passes through the tissue. From these regional signal changes, it is possible to calculate parameters that include the rate of perfusion by understanding the fractional volume of gadolinium in the extravascular–extracellular space compared with the fractional volume of gadolinium in the plasma. See, Essig et al., 2013, “Perfusion MRI: The Five Most Frequently Asked Technical Questions,” Am. J. Roentgenol.200, pp.24–34, which is hereby incorporated by reference. These differences in T1 and T2 * signal are due to the T1 and T2 shortening effects of gadolinium, resulting in high T1 and low T2 or T2 * signal, respectively. A third technique, known as arterial spin labelling, is a perfusion sequence that does not require intravenous contrast administration. It harnesses the ability of the MRI to selectively label inflowing arterial blood and monitor tissue perfusion. In so doing, the protons in flowing arterial blood act as endogenous contrast to calculate bolus parameters such as cerebral blood flow. See, Petcharunpaisan et al., 2010, “Arterial spin labeling in neuroimaging,” World J. Radiol.2, pp.384–398; and Telischak et al., 2015, “Arterial spin labeling MRI: Clinical applications in DB2/ 47566363.1 27 Attorney Ref. No.: 118547-5008-WO the brain,” J. Magn. Reson. Imaging 41, pp.1165–1180, each of which is hereby incorporated by reference. [00104] In some embodiments, computed tomography is used to obtain the radiographic modality finding. See, Vidhya et al., 2021, “Automated Detection and Screening of Traumatic Brain Injury (TBI) Using Computed Tomography Images: A Comprehensive Review and Future Perspectives,” . Int. J. Environ. Res. Public Health 18, p.6499, which is hereby incorporated by reference. In some embodiments the computed tomography is non- contrast computed tomography. In some embodiments, the computed tomography is positron emission tomography (PET) or single-photon emission computed tomography (SPECT). [00105] Referring to block 910, in some embodiments, the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage (SAH). In some embodiments the radiographic modality finding of SAH is determined using computed tomography. See, for example, Alwageed, 2022, “Detection of Subarachnoid Hemorrhage in Computed Tomography Using Association Rules Mining,” Computational Intelligence and Neuroscience Volume 2022, Article ID 1133819; and Marcolini and Hine, 2019, “Approach to the Diagnosis and Management of Subarachnoid Hemorrhage,” West J Emerg Med 20(2): pp.203-211, each of which is hereby incorporated by reference. In some embodiments, the radiographic modality finding is a binary indication as to whether or not the mTBI subject has a subarachnoid hemorrhage (SAH). [00106] Referring to block 912, in some embodiments, the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage (IVH). In some embodiments the radiographic modality finding of an IVH is determined using computed tomography. See, for example, Nawabi et al., “Non‑contrast computed tomography features predict intraventricular hemorrhage growth,” European Radiology 33, pp.7807-7817, which is hereby incorporated by reference. In some embodiments, the radiographic modality finding is a binary indication as to whether or not the mTBI subject has an IVH. [00107] Referring to block 914, in some embodiments, the radiographic modality finding of the mTBI subject consists of (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. DB2/ 47566363.1 28 Attorney Ref. No.: 118547-5008-WO [00108] Referring to block 916, in some embodiments, the radiographic modality finding of the mTBI subject comprises a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture. In some such embodiments, the radiographic modality is computed tomography. [00109] An epidural hematoma refers to a medical condition where bleeding occurs between the skull and the outer layer of the membrane covering the brain (dura mater). This condition typically arises due to trauma or injury to the head, leading to the rupture of blood vessels in the epidural space. CT imaging can be used to diagnose epidural hematomas because it provides detailed cross-sectional images of the brain and surrounding structures. In a CT scan, an epidural hematoma appears as a lens-shaped or biconvex mass of blood located between the inner surface of the skull and the dura mater. The hematoma may exert pressure on the brain, leading to symptoms such as headache, dizziness, nausea, vomiting, confusion, weakness, or loss of consciousness. [00110] A computed tomography (CT) Marshall score, often referred to simply as the Marshall score or classification, is a system used to categorize traumatic brain injury (TBI) based on findings from CT scans of the head. See, for example, Mohammadifard et al., 2008, “Marshall and Rotterdam Computed Tomography scores in predicting early deaths after brain trauma,” Eur J Transl Myol 28(3), pp.265-273, which is hereby incorporated by reference. [00111] An edema refers to an accumulation of fluid in the brain tissue. It can be discovered, for example, using computed tomography, magnetic resonance imaging, diffusion-weighted imaging, fluid-attenuated inversion recovery (FLAIR) imaging, positron emission tomography, single photon emission computed tomography, and ultrasound. [00112] A contusion refers to a bruise on the brain caused by trauma, such as a blow to the head or rapid deceleration injury. A contusion can be discovered, for example, using computed tomography, magnetic resonance imaging, diffusion-weighted imaging, fluid- attenuated inversion recovery (FLAIR) imaging, gradient echo (GRE) imaging, positron emission tomography, and single photon emission computed tomography. [00113] A subdural hematoma refers to a type of bleeding that occurs between the layers of tissue surrounding the brain. It typically develops as a result of trauma to the head, causing DB2/ 47566363.1 29 Attorney Ref. No.: 118547-5008-WO blood vessels to rupture and blood to accumulate in the space between the dura mater (the outermost layer covering the brain) and the arachnoid mater (the middle layer). Subdural hematomas can vary in severity and can be classified based on the time elapsed since the initial bleeding. A subacute subdural hematoma indicates that the hematoma is in an intermediate stage of development, typically between 3 days to 3 weeks after the initial bleeding event. During this stage, the hematoma may appear more organized and may have different radiological characteristics compared to an acute hematoma. A subacute subdural hematoma can be discovered, for example, using computed tomography, magnetic resonance imaging, ultrasound, angiography, positron emission tomography, and single photon emission computed tomography. [00114] An extraaxial hematoma refers to a hematoma that occurs outside the brain tissue, while "hematoma" refers to a collection of blood. Extra-axial hematomas can occur in different locations within the skull, typically between the layers of tissue that cover the brain. Common types of extra-axial hematomas include epidural hematomas, subdural hematomas, and subarachnoid hemorrhages. Each of these types of hematomas has distinct characteristics and can result from different causes, such as trauma. An extraaxial hematoma can be discovered, for example, using computed tomography, magnetic resonance imaging, ultrasound, angiography, positron emission tomography, and single photon emission computed tomography. [00115] CT is particularly effective in detecting skull fractures because it provides high- resolution images of the bones and surrounding soft tissues. It can detect even subtle fractures that may not be visible on conventional X-rays. Additionally, CT can capture multiple views and angles of the skull, allowing for a comprehensive evaluation of the extent and distribution of fractures. [00116] A Marshall 234 score is derived from the Marshall classification of traumatic brain injury (MCTC), a CT-scan derived metric using only a few features and has been shown to predict outcome in patients with traumatic brain injury. This system was first published in 1992 and remains one of the most commonly used systems for grading acute traumatic brain injury on the basis of CT findings. See, Mahadewa et al., 2018, “Modified Revised Trauma– Marshall score as a proposed tool in predicting the outcome of moderate and severe traumatic brain injury,” Emerg Med.10, pp.135-139, which is hereby incorporated by reference. In accordance with the Marshall scoring: DB2/ 47566363.1 30 Attorney Ref. No.: 118547-5008-WO [00117] diffuse injury I (no visible pathology) a. no visible intracranial pathology [00118] diffuse injury II a. midline shift of 0 to 5 mm b. basal cisterns remain visible c. no high or mixed density lesions >25 cm3 [00119] diffuse injury III (swelling) a. midline shift of 0 to 5 mm b. basal cisterns compressed or completely effaced c. no high or mixed density lesions >25 cm3 [00120] diffuse injury IV (shift) a. midline shift >5 mm b. no high or mixed density lesions >25 cm3 [00121] evacuated mass lesion V a. any lesion evacuated surgically [00122] non-evacuated mass lesion VI a. high or mixed density lesions >25 cm3 b. not surgically evacuated [00123] The Marshall 234 score is a binary indication of whether or not an mTBI subject has a Marshall 2, 3, or 4 category score. The Marshall 56 score is a binary indication of whether or not an mTBI subject has a Marshall 5 or 6 category score. The first group for patients with a Marshall group of 1 is omitted to avoid issues of multicollinearity, since it is implicitly defined by the absence of the other two categories, e.g, when both binary variables for Marshall 234 and Marshall 56 are 0, it indicates that the mTBI subject belongs to Marshall group 1. [00124] Referring to block 918, in some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury. [00125] In some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two weeks of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two weeks of incurring the traumatic brain injury. DB2/ 47566363.1 31 Attorney Ref. No.: 118547-5008-WO [00126] Referring to block 920, in some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury. [00127] In some embodiments, the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 days of incurring the traumatic brain injury. [00128] Referring to block 922, an indication is provided, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit. [00129] In some embodiments, a mTBI subject incurs a chronic functional outcome deficit when they have a Glasgow Outcome Scale-Extended (GOS-E) score of less than or equal to 7, twelve months after incurring the initial traumatic brain injury. [00130] In some embodiments, a mTBI subject incurs a chronic functional outcome deficit when they have a GOSE score of less than or equal to 7, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 months after incurring the initial traumatic brain injury. [00131] In some embodiments, a mTBI subject incurs a chronic functional outcome deficit when they have a GOS-E score of less than or equal to 6, twelve months after incurring the initial traumatic brain injury. [00132] In some embodiments, a mTBI subject incurs a chronic functional outcome deficit when they have a GOS-E score of less than or equal to 6, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 months after incurring the initial traumatic brain injury. [00133] For GOS-E scores, see McMahon, et al., 2014, “Symptomatology and Functional Outcome in Mild Traumatic Brain Injury: Results from the Prospective TRACK-TBI Study,” Journal of Neurotrauma 31, pp.26-33, which is hereby incorporated by reference. In some embodiments, an equivalent test is used to establish that a subject incurs a chronic functional outcome deficit. [00134] GOS-E is a commonly used scale to assess the functional outcome of mTBI subjects following a traumatic brain injury. GOS-E provides a structured way of categorizing mTBI DB2/ 47566363.1 32 Attorney Ref. No.: 118547-5008-WO subjects into different levels of recovery based on their level of disability and ability to function in daily life. The GOS-E scores range from 1 to 8, with 1 representing death and 8 representing upper good recovery with the resumption of normal activities. The following is a breakdown of Glasgow Outcome Scale-Extended scores: [00135] 1. Dead: The patient has died. [00136] 2. Vegetative State: The patient is unconscious and shows no evidence of awareness of self or environment. [00137] 3. Lower Severe Disability: The patient is conscious but requires constant nursing care due to severe neurological impairment. They are unable to live independently. [00138] 4. Upper Severe Disability: The patient is conscious but requires assistance with daily living activities. They are unable to live independently but may be able to follow simple commands. [00139] 5. Lower Moderate Disability: The patient is able to live independently but may require some assistance with certain tasks or have some cognitive or physical deficits. [00140] 6. Upper Moderate Disability: The patient is able to live independently with minimal assistance but may have some persisting symptoms or disabilities that affect daily life. [00141] 7. Lower Good Recovery: The patient has made a good recovery and is able to resume normal activities but may have some minor impairments or limitations. [00142] 8. Upper Good Recovery: The patient has made a full recovery and has resumed normal activities without any significant impairments. [00143] Referring to block 924, in some embodiments, the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit. [00144] Referring to block 926, in some embodiments, the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit. [00145] Referring to block 928, in some embodiments, the method further comprises obtaining a baseline heart rate, a diastolic blood pressure, a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject. The indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based DB2/ 47566363.1 33 Attorney Ref. No.: 118547-5008-WO on the baseline heart rate, the diastolic blood pressure, the respiratory rate, the temperature, the systolic blood pressure, or any combination thereof, of the mTBI subject. [00146] Referring to block 930, in some embodiments, the method further comprises obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject. The indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject. [00147] Referring to block 932, in some embodiments, the providing the indication comprises: inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit. [00148] Referring to block 934, in some embodiments, the model is an Elastic Net model with a multinomial likelihood function. [00149] Referring to block 936, in some embodiments, the model is a nonnegative matrix factorization model. [00150] Referring to block 938, in some embodiments, the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof. [00151] Referring to block 940, in some embodiments, the inputting further comprises inputting one or more covariates into the model. DB2/ 47566363.1 34 Attorney Ref. No.: 118547-5008-WO [00152] Referring to block 942, in some embodiments, the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof. Moreover, other covariates, alone or in any combination can be inputted into the model, such as whether or not the subject has a history of substance abuse or an extent of substance abuse, whether or not the subject has an alteration in the hypothalamic-pituitary- adrenal (HPA) axis function or abnormalities in brain structure and function (e.g., hippocampal volume reduction) or an extent to such alterations or abnormalities, whether or not the subject suffers from depression or an extent of such depression, or whether or not the subject has MDD. [00153] Referring to block 944, in some embodiments, the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises treating the mTBI subject based on the indication. [00154] Referring to block 946, in some embodiments, the treating the mTBI subject based on the indication comprises administering psychotherapy to the mTBI subject. [00155] Psychotherapy can be beneficial for individuals with mild traumatic brain injury (mTBI) to address cognitive, emotional, and behavioral challenges that may arise following the injury. Cognitive behavioral therapy (CBT) is one form of psychotherapy. CBT is a structured form of therapy that focuses on identifying and changing negative thought patterns and behaviors. It can help mTBI subjects manage symptoms such as anxiety, depression, and stress, as well as address cognitive difficulties such as memory problems and difficulty concentrating. See, for example, Miller and Mittenberg, 2010, “Brief Cognitive Behavioral Interventions in Mild Traumatic Brain Injury,” Applied Neuropsychology 5:4, pp.172-183, which is hereby incorporated by reference. [00156] Psychoeducation is another form of psychotherapy. Psychoeducation involves providing information and education about mTBI, its symptoms, and its effects on cognitive and emotional functioning. This can help mTBI subjects better understand their condition and learn strategies for coping with and managing symptoms. See, for example, Caplain et al, 2019, “Efficacy of Psychoeducation and Cognitive Rehabilitation After Mild Traumatic Brain Injury for Preventing Post-concussional Syndrome in Individuals with High Risk of DB2/ 47566363.1 35 Attorney Ref. No.: 118547-5008-WO Poor Prognosis: A Randomized Clinical Trial,” Neurotram 10, DOI=10.3389/fneur.2019.00929, which is hereby incorporated by reference. [00157] Neuropsychological rehabilitation is another form of psychotherapy. Neuropsychological rehabilitation involves using cognitive training exercises and strategies to help improve cognitive functioning following mTBI. This may include exercises to improve memory, attention, and problem-solving skills, as well as compensatory strategies to help mTBI subjects work around areas of cognitive difficulty. See, for example, 2009, “Neuropsychological rehabilitation of mild traumatic brain injury,” Brain Injury 10:4, pp. 277-286, which is hereby incorporated by reference. [00158] Mindfulness-based interventions is another form of psychotherapy. Mindfulness- based interventions, such as mindfulness-based stress reduction (MBSR) or mindfulness- based cognitive therapy (MBCT), can help mTBI subjects reduce stress, improve mood, and enhance overall well-being by teaching them to cultivate present-moment awareness and acceptance. See, for example, Lovette et al., 2022, ‘“Hidden gains”? Measuring the impact of mindfulness-based interventions for people with mild traumatic brain injury: a scoping review,’ Brain Injury 36:9, pp.1059-1070, which is hereby incorporated by reference. [00159] Supportive therapy is another form of psychotherapy. Supportive therapy involves providing emotional support and validation to mTBI subjects as they navigate the challenges of living with their injury. This type of therapy can help mTBI subjects feel understood, accepted, and less alone in their struggles. [00160] Family therapy another form of psychotherapy. Family therapy can be helpful for mTBI subjects and their families to address the impact of the injury on family dynamics and relationships. Family therapy can help families learn effective communication strategies, resolve conflicts, and provide support to one another during the recovery process. See, for example, Dausch and Sailiman, 2009, “Use of family focused therapy in rehabilitation for veterans with traumatic brain injury,” Rehabilitation Psychology 54(3), pp.279-287, which is hereby incorporated by reference. [00161] Referring to block 948, in some embodiments, the treating the mTBI subject based on the indication comprises administering physical therapy to the mTBI subject. Physical therapy for mTBI subjects focuses on addressing physical symptoms and restoring function that may have been affected by the injury. Some forms of physical therapy commonly administered to mTBI subjects are described below. DB2/ 47566363.1 36 Attorney Ref. No.: 118547-5008-WO [00162] Vestibular rehabilitation therapy (VRT) is a form of physical therapy that can be administered to mTBI subjects. Many mTBI subjects experience vestibular dysfunction, which can cause symptoms such as dizziness, vertigo, and imbalance. VRT is a specialized form of physical therapy designed to address these symptoms by using exercises and techniques to improve balance, coordination, and proprioception. See, for example, Hofferx and Balabany, 2011, “Vestibular rehabilitation after mild traumatic brain injury with vestibular pathology,” NeuroRehabilitation 29(2), pp.167-171, which is hereby incorporated by reference. [00163] Oculomotor rehabilitation is another form of physical therapy that can be administered to mTBI subjects. Some mTBI subjects may experience visual disturbances and difficulties with eye movements following their injury. Oculomotor rehabilitation involves exercises and techniques to improve eye tracking, fixation, convergence, and other aspects of visual function. See, for example, Thiagarajan and Ciuffreda, 2013, “Effect of oculomotor rehabilitation on vergence responsivity in mild traumatic brain injury,” JRRD 50(9), pp. 1223-1240, which is hereby incorporated by reference. [00164] Balance training is another form of physical therapy that can be administered to mTBI subjects. Balance training exercises are commonly used in physical therapy to help mTBI subjects improve their balance and stability. These exercises may include standing on one leg, walking on uneven surfaces, and performing dynamic balance activities to challenge the vestibular and proprioceptive systems. See, for example, Kakade and Kanase, 2020, “Effect of Multidimensional Exercise Program for Improving Balance in Traumatic Brain Injury Patients,” Medico-legal Update, July-September 20(3), pp.155-161, which is hereby incorporated by reference. [00165] Aerobic exercise is another form of physical therapy that can be administered to mTBI subjects. Aerobic exercise, such as walking, cycling, or swimming, can be beneficial for mTBI subjects to improve cardiovascular fitness, endurance, and overall physical health. Aerobic exercise has also been shown to have positive effects on mood and cognitive function. See, for example, Chin et al., 2015, “Improved Cognitive Performance Following Aerobic Exercise Training in People With Traumatic Brain Injury,” Archives of Physical Medicine and Rehabilitation 96(4), pp.754-759, which is hereby incorporated by reference. [00166] Strengthening exercises are another form of physical therapy that can be administered to mTBI subjects. Strengthening exercises may be prescribed to mTBI subjects DB2/ 47566363.1 37 Attorney Ref. No.: 118547-5008-WO to help improve muscle strength, endurance, and functional capacity. These exercises may target specific muscle groups that have been weakened or deconditioned as a result of the injury. See, for example, Alarie et al., 2022, “Physical activity interventions in rehabilitation programs for outpatients with mild traumatic brain injury,” Research Quarterly for Exercise and Sport 93(4), pp.851-860, which is hereby incorporated by reference. [00167] Gait training is another form of physical therapy that can be administered to mTBI subjects. Gait training involves working on walking patterns and mechanics to improve walking speed, stride length, and overall gait efficiency. This may include exercises to improve posture, balance, and coordination while walking. See, for example, Alashram, 2019, “Optimizing gait ability after task-oriented circuit class training in posttraumatic brain injury: a case report,” Indian J Phys Med Rehabil.30(4), pp.112-116, which is hereby incorporated by reference. [00168] Manual therapy is another form of physical therapy that can be administered to mTBI subjects. Manual therapy techniques, such as massage, joint mobilization, and soft tissue mobilization, may be used by physical therapists to help alleviate muscle tension, reduce pain, and improve mobility in mTBI subjects. See, for example, Kane et al., 2019, “Physical therapy management of adults with mild traumatic brain injury,” in Seminars in Speech and Language 40(01), pp.36-47, Thieme Medical Publishers, which is hereby incorporated by reference. [00169] Referring to block 950, in some embodiments, the treating the mTBI subject based on the indication comprises administering speech therapy to the mTBI subject. Speech therapy for mTBI subjects focuses on addressing communication difficulties that may arise as a result of the injury. [00170] One form of speech therapy for mTBI subjects is cognitive-communicative therapy. Cognitive-communicative therapy targets cognitive processes involved in communication, such as attention, memory, executive function, and problem-solving. This type of therapy aims to improve the ability to understand and use language effectively in various communication situations. See, for example, MacLennan, 2012, “Cognitive-communication rehabilitation for combat-related mild traumatic brain injury,” Journal of rehabilitation research and development 49(7):XI, which is hereby incorporated by reference. [00171] Another form of speech therapy for mTBI subjects is language therapy. Language therapy focuses on improving specific language skills that may be affected by mTBI, such as DB2/ 47566363.1 38 Attorney Ref. No.: 118547-5008-WO comprehension, expression, vocabulary, and grammar. Therapy may involve exercises to improve word finding, sentence structure, and overall language fluency. See, for example, Hardin and Kelly, 2019, “The role of speech-language pathology in an interdisciplinary care model for persistent symptomatology of mild traumatic brain injury,” in Seminars in Speech and Language 40(01), pp.065-078, Thieme Medical Publishers, which is hereby incorporated by reference. [00172] Another form of speech therapy for mTBI subjects is speech sound production therapy. Some mTBI subjects may experience difficulties with speech sound production, such as articulation, phonation, and resonance. Speech sound production therapy involves exercises and techniques to improve speech clarity, accuracy, and intelligibility. [00173] Another form of speech therapy for mTBI subjects is voice therapy. Voice therapy may be recommended for mTBI subjects who experience changes in voice quality, pitch, volume, or resonance following their injury. Therapy may include exercises to improve vocal fold function, breath support, and vocal resonance. [00174] Another form of speech therapy for mTBI subjects is fluency therapy. Fluency therapy is aimed at improving speech fluency and reducing stuttering or other disfluencies that may occur as a result of mTBI. Therapy may involve strategies to increase relaxation, slow speech rate, and enhance smoothness of speech production. [00175] Another form of speech therapy for mTBI subjects is social communication skills training. Social communication skills training focuses on improving pragmatic language skills, such as turn-taking, topic maintenance, conversational repair, and nonverbal communication. This type of therapy may involve role-playing, video modeling, and structured social communication activities. See, for example, Dahlberg et al., 2006, “Social communication skills training after traumatic brain injury,” The Journal of Head Trauma Rehabilitation 21(5), p.425, which is hereby incorporated by reference. [00176] Another form of speech therapy for mTBI subjects is augmentative and alternative communication (AAC). For mTBI subjects with severe communication difficulties or speech impairments that persist following mTBI, AAC systems and devices may be used to supplement or replace verbal communication. Speech therapists can assess the individual's communication needs and provide training on how to use AAC systems effectively. [00177] Referring to block 952, in some embodiments, the treating the mTBI subject based on the indication comprises administering occupational therapy to the mTBI subject. DB2/ 47566363.1 39 Attorney Ref. No.: 118547-5008-WO Occupational therapy for mTBI subjects focuses on improving functional abilities and facilitating a return to daily activities. [00178] One form of occupational therapy for mTBI subjects is activities of daily living (ADL) Training. In ADL training, medical practitioners such as occupational therapists work with mTBI subjects to improve their ability to perform basic self-care tasks, such as bathing, dressing, grooming, toileting, and feeding. This may involve breaking down tasks into smaller steps, using adaptive equipment, and practicing skills in a structured environment. See, for example, Moon and Jeon, 2019, “Effects of Virtual Reality-Based Activities of Daily Living Training on Activities of Daily Living and Rehabilitative Motivation in Patients with Traumatic Brain Injury: A Pilot Study,” Therapeutic Science for Rehabilitation.8(4), pp.41- 51, which is hereby incorporated by reference. [00179] Another form of occupational therapy for mTBI subjects is instrumental activities of daily living (IADL) training. In IADL training, medical practitioners such as occupational therapists help mTBI subjects develop or regain skills needed for more complex daily activities, such as meal preparation, household chores, managing finances, and using transportation. This may include cognitive strategies, environmental modifications, and task- specific training. See, for example, Ertas-Spantgar et al., 2024, “Guiding patients with traumatic brain injury through the instrumental activities of daily living with the RehaGoal App: a feasibility study,” Disability and Rehabilitation: Assistive Technology19(2), pp.254- 65, which is hereby incorporated by reference. [00180] Another form of occupational therapy for mTBI subjects is cognitive rehabilitation. Cognitive rehabilitation focuses on improving cognitive functions that may be affected by mTBI, such as attention, memory, executive function, and problem-solving. Medical practitioners such as occupational therapists may use various techniques, such as cognitive exercises, compensatory strategies, and environmental adaptations, to help mTBI patients manage cognitive difficulties and perform daily activities more independently. See, for example, Allen, 2019, “Cognitive rehabilitation for mild traumatic brain injury (mTBI),” in Neurosensory disorders in mild traumatic brain injury, pp.357-379, Academic Press, which is hereby incorporated by reference. [00181] Another form of occupational therapy for mTBI subjects is sensory integration therapy. Some mTBI subjects may experience sensory processing difficulties, such as hypersensitivity or hyposensitivity to sensory input. Sensory integration therapy involves DB2/ 47566363.1 40 Attorney Ref. No.: 118547-5008-WO activities and exercises designed to help regulate sensory processing and improve tolerance to sensory stimuli, which can enhance participation in daily activities. [00182] Another form of occupational therapy for mTBI subjects is fatigue management. Fatigue is a common symptom of mTBI that can significantly impact daily functioning. Medical practitioners such as occupational therapists help mTBI subjects manage fatigue by teaching energy conservation techniques, scheduling rest breaks, prioritizing tasks, and modifying activities to reduce physical and cognitive exertion. See, for example, Xu et al.. 2017, “Complementary and alternative interventions for fatigue management after traumatic brain injury: a systematic review,” Therapeutic Advances in Neurological Disorders 10(5), pp.229-239, which is hereby incorporated by reference. [00183] Another form of occupational therapy for mTBI subjects is community reintegration. Medical practitioners, such as occupational therapists support mTBI subjects in returning to meaningful activities and roles in their community, such as work, school, volunteering, and social activities. This may involve vocational counseling, job coaching, school reintegration services, and leisure exploration to help mTBI subjects identify and pursue meaningful activities post-injury. See, for example, Belanger, 2018, “Community Reintegration,” Rehabilitation After Traumatic Brain Injury, Eapin and Cifu, eds. p.255, Elsevier, Saint Louis, Missouri, which is hereby incorporated by reference. [00184] Referring to block 954, in some embodiments, the treating the mTBI subject based on the indication comprises administering melatonin or zolipidem to the mTBI subject. Insomnia is highly prevalent within the mTBI population and is a subtle, frequently persistent complaint that often goes undiagnosed. For mTBI subjects, problems with sleep can compromise the recovery process and impede social reintegration. Thus, in some embodiments the treating the mTBI subject comprises administering to the subject melatonin or zolipidem. See, for example, Zhou and Greenwald, 2018, “Update on Insomnia after Mild Traumatic Brain Injury,” Brain Sci.8(12), p.223, which is hereby incorporated by reference. [00185] Referring to block 956, in some embodiments, the treating the mTBI subject based on the indication comprises administering a selective serotonin reuptake inhibitor (SSRI) to the mTBI subject. SSRIs are a class of antidepressant agents that inhibit the reuptake of serotonin by monoamine transporters in the presynaptic cell, allowing for increased availability of serotonin in the synaptic cleft and increased/repeated stimulation of serotonergic postsynaptic receptors, leading to increased synaptic signaling. SSRIs increase DB2/ 47566363.1 41 Attorney Ref. No.: 118547-5008-WO extracellular serotonin (5-HT), activating the seven types of 5-HT receptors and many subtypes of each class, which can also be directly stimulated by the SSRI. Non-limiting examples of SSRIs included, but are not limited to Sertraline, Citalopram, Fluoxetine, Paroxetine, Sertraline, Prednisone, and Carbamazepine. See, for example, Yue et al., 2017, “Selective serotonin reuptake inhibitors for treating neurocognitive and neuropsychiatric disorders following traumatic brain injury: an evaluation of current evidence,” Brain sciences 7(8), p.93, which is hereby incorporated by reference. [00186] Referring to block 958, in some embodiments, the treating the mTBI subject based on the indication comprises administering methylphenidate or amphetamine salt to the mTBI subject. See, for example, Coris et al., 2022, “Stimulant therapy utilization for neurocognitive deficits in mild traumatic brain injury,” Sports Health 14(4), pp.538-548, which is hereby incorporated by reference. [00187] Referring to block 960, in some embodiments, the treating the mTBI subject based on the indication comprises administering cyclobenzaprine to the mTBI subject. Cyclobenzaprine is a muscle relaxant medication commonly prescribed for the relief of muscle spasms and associated pain. Cyclobenzaprine may be prescribed in certain situations for specific symptoms commonly experienced after mild traumatic brain injury, such as to treat muscle spasms, to provide pain relief, or to improve sleep. [00188] After a traumatic brain injury, particularly if there has been associated head or neck trauma, muscle spasms can occur as a result of muscle strain or injury. Cyclobenzaprine can help reduce muscle spasms, which may alleviate associated discomfort and improve mobility. [00189] Headaches and neck pain are common symptoms following mTBI, often due to muscle tension or strain. Cyclobenzaprine can help alleviate these symptoms by relaxing muscles and reducing associated pain. [00190] Cyclobenzaprine has sedative effects, and it is sometimes prescribed to help individuals with mTBI who are experiencing sleep disturbances. Improving sleep quality can be important for overall recovery from mTBI, as adequate rest supports brain healing and cognitive function. [00191] Cyclobenzaprine may be prescribed as part of an adjunctive therapy regimen alongside other medications or therapies to manage symptoms such as headaches, neck pain, and muscle stiffness associated with mTBI. DB2/ 47566363.1 42 Attorney Ref. No.: 118547-5008-WO [00192] Referring to block 962, in some embodiments, the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises adjusting a treatment regime of the mTBI subject based on the indication. [00193] Referring to block 964, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a frequency or duration of psychotherapy for the mTBI subject. Examples of psychotherapy that may be adjusted in accordance with block 964 are described in block 946. [00194] Referring to block 966, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a frequency or a duration of physical therapy for the mTBI subject. Examples of physical therapy that may be adjusted in accordance with block 966 are described in block 948. [00195] Referring to block 968, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a frequency or a duration of speech therapy for the mTBI subject. Examples of speech therapy that may be adjusted in accordance with block 968 are described in block 950. [00196] Referring to block 970, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a frequency or a duration of occupational therapy for the mTBI subject. Examples of occupational therapy that may be adjusted in accordance with block 970 are described in block 952. [00197] Referring to block 972, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a dosage of melatonin or zolipidem given to the mTBI subject or increasing or decreasing a length of time the melatonin or zolipidem is given to the mTBI subject. Melatonin or zolipidem treatment regimens that may be adjusted in accordance with block 972 are described in block 954. [00198] Referring to block 974, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a dosage of a selective serotonin reuptake inhibitor given to the mTBI subject or increasing or decreasing a length of time the selective serotonin reuptake inhibitor is given to the mTBI subject. Selective serotonin reuptake inhibitor treatment regimens that may be adjusted in accordance with block 974 are described in block 956. [00199] Referring to block 976, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a dosage of methylphenidate or amphetamine salt given to DB2/ 47566363.1 43 Attorney Ref. No.: 118547-5008-WO the mTBI subject or increasing or decreasing a length of time the methylphenidate or amphetamine salt is given to the mTBI subject. Methylphenidate or amphetamine salt treatment regimens that may be adjusted in accordance with block 976 are described in block 958. [00200] Referring to block 978, in some embodiments, the adjusting the treatment regime comprises increasing or decreasing a dosage of cyclobenzaprine give to the mTBI subject or increasing or decreasing a length of time the cyclobenzaprine is given to the mTBI subject. Cyclobenzaprine treatment regimens that may be adjusted in accordance with block 978 are described in block 960. [00201] Example 1. [00202] Prognostic biomarkers could potentially advance precision therapeutics and provide insight into the causal underpinnings of chronic mTBI symptoms. In this example, clusters of mTBI patients were characterized based on chronic psychiatric and functional symptom profiles 12 months after injury, and to predict cluster membership from baseline biological data. Data from the TRACK-TBI cohort (total N = 1377 patients) was used, clustering patients with Nonnegative Matrix Factorization of chronic symptoms, and machine learning models were built using vital signs, lab work, derived computed tomography and MRI findings, and blood biomarkers from the time of injury in order to predict 12 month cluster assignment. The data were well characterized by five clusters, corresponding to chronic Functional Outcome, Post-Traumatic Stress Disorder (PTSD), Depression, Sleep Disturbance, and Life Satisfaction subtypes. Data indicate that 12 month Functional Outcome subtypes can be well predicted (mean out-of-sample AUC > 0.7) from baseline computed tomography and high-sensitivity C-reactive Protein (hsCRP) features, and that re- experiencing and avoidance PTSD sub-types are predictable with moderate accuracy (mean out-of-sample AUC > 0.6) from Neuron Specific Enolase (NSE) and Sodium levels. These results demonstrate that blood- and CT-based markers are successful predictors of long-term functional outcome and could contribute towards prognostic biomarkers. [00203] This exampled aimed to advance the understanding and management of mTBI through two primary aims: [00204] First, to identify and characterize clusters of mTBI patients based on a variety of chronic symptoms prevalent at 12 months. This classification was achieved using unsupervised learning techniques to analyze data encompassing measures across psychiatric DB2/ 47566363.1 44 Attorney Ref. No.: 118547-5008-WO and daily living scales. By identifying these clusters, light was shed on the diverse trajectories of mTBI recovery, providing additional understanding of its long-term impact. [00205] Secondly, to build machine learning based models capable of predicting the 12- month outcome cluster for individual patients, using only baseline biological data. This goal involved the integration of a wide array of patient data, combining vital signs, laboratory results, Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scans, and levels of three candidate liquid biopsy-based biomarkers (hsCRP, Neuron Specific Enolase (NSE), and S100 calcium-binding protein B (S100B)). Data from TRACK-TBI [29], a multi- center study that followed patients longitudinally from their emergency room visits, was used. [00206] The methodology used in this example is distinct in its comprehensive utilization of unsupervised machine learning to dissect the heterogeneity of 12-month outcomes across multiple domains. By integrating baseline data from various biological modalities, this example not only sought to characterize the complexity of mTBI recovery but also employed machine learning techniques to directly predict chronic sub-types in individual patients, rather than focusing on population-level statistical associations. [00207] Clinical data 12 months after injury. Data from TRACK-TBI [29] were downloaded from the FITBIR [30] database. Because symptom heterogeneity following a mild TBI was of interest, only patients were included in this report. mTBI patients with Glasgow Coma Scale (GCS) [31] score in the range of 13-15 were included, whether computed tomography findings were positive or negative. For psychiatric symptom and functional outcome measures, data was combined from Patient Health Questionnaire-9 (PHQ-9) [32], Glasgow Outcome Scale Extended (GOS-E) [33], Insomnia Severity Index (ISI) [34], Satisfaction With Life Scale (SWLS) [35], Mayo-Portland Adaptability Inventory (MPAI) [36], and the Brief Symptom Inventory (BSI) [37]. Because chronic outcomes were of interest, data from the 12 month follow-up visit was selected (≥329 days, 367 ± 16 [mean ± SD] days) leading to a total N of 1467. [00208] PHQ-9 is a self-report, standardized depression rating scale is used for screening, diagnosing, and monitoring treatment response for major depressive disorder (MDD). See Kroenke et al., 2003, “The Patient Health Questionnaire-2: validity of a two-item depression screener,” Med Care 41(11), pp.1284-92, which is hereby incorporated by reference. The PHQ-9 uses 9 items corresponding to the DSM-5 criteria for MDD and also assesses for DB2/ 47566363.1 45 Attorney Ref. No.: 118547-5008-WO psychosocial impairment. The PHQ-9 scores 0 to 27, with scores of equal to or more than 10, indicate a possible MDD. See Bains and Abdijadid, Major Depressive Disorder, [Updated 2023 Apr 10], In: StatPearls [Internet], Treasure Island (FL): StatPearls Publishing; 2024 Jan- . Available from: https://www.ncbi.nlm.nih.gov/books/NBK559078/, which is hereby incorporated by reference. [00209] Observations missing 30% or more items across all scales were excluded. To deal with missingness in the remaining participants, missing data was imputed using Predictive Mean Matching (PMM) [38] in the MICE package [39] in R. During imputation, scale scores were treated as ordered factors. After removal and imputation, a total N of 1377 patients were included in 12 month symptom clustering. Imputed scales were normalized using the min-max function before clustering. [00210] Nonnegative Matrix Factorization of chronic symptoms. [00211] In order to find subtypes of patients with similar profiles of chronic symptoms measured 12 months after injury, Nonnegative Matrix Factorization (NMF) [40,41] was used, implemented in the NMF package in R [42]. While this technique is most commonly used for dimensionality reduction onto a continuous latent space, it is known to produce sparse and meaningful latent dimensions, and is thus also appropriate for fuzzy clustering [43–45]. Observations can be assigned to multiple communities which allows assessment of uncertainty of community assignments. In addition, the fact that the clinical scales measured at 12 months are bounded at 0 makes NMF an appealing framework for clustering in this setting. [00212] NMF provides a low-rank approximation of a non-negative matrix ^^^^ ∈ ℝ ^ ≥^^^ 0× ^^^^, decomposing it into the product of two matrices: an ^^^^ × ^^^^ basis matrix ^^^^ and a
Figure imgf000048_0001
coefficient matrix ^^^^: ^^^^ ≈ ^^^^ ^^^^ [00213] For the 12 month clinical data described above, X is a ^^^^ × ^^^^ matrix of clinical symptom scores, where d is the number of items and n is the number of participants. k is the rank of the approximation: the number of dimensions, and in this example the number of clusters. W provides the weighting of each scale item in each of the k dimensions of the lower rank approximation, and H provides the coordinates of each patient in the basis W. Cluster membership is derived for each participant from the coefficients H by selecting the dimension with the highest H value. DB2/ 47566363.1 46 Attorney Ref. No.: 118547-5008-WO [00214] The NMF model was fit using an objective function derived from [46], which minimizes the Kullback-Leibler divergence between WH and X. In order to select the rank of the factorization, two sets of models ranging from 2 to 10 factors were estimated. One set of models made use of the data X, while the other set of models was fit to data randomized across columns (scale items). A number of statistics measuring NMF fit for real and randomized data are plotted in Figure 2A. Each model was run 5 times, and the average of each fit statistic is shown for each rank. In particular, the silhouette score was of interest, as this is a measure of goodness of clustering [47]. In the case of NMF, silhouette scores can be generated for either basis or coefficient matrices. Because the goal is to cluster patients (as opposed to scale items), the coefficient silhouette was focused on. In Figure 2A), the cophenetic chart plots consensus, while the dispersion, evar, residuals, and rss plot best ft. In Figure 2A), lines with circles represent random data while lines with triangles represent real data. In Figure 2A, line 202 is consensus type, line 204 is coefficients type, line 206 is coefficients type, line 208 is basis type, line 210 is consensus type, line 212 is basis type, line 214 is coefficients type, line 216 is coefficients type, line 218 is basis type, and line 220 is basis type. Figure 2B shows the difference in coefficient silhouette score between real versus random data clustering. For subsequent analyses, k was set to 5 clusters, as this produced the largest difference in silhouette score, and was at or near the highest difference for most other measures. [00215] It is worth noting that no claim that 5 is the true or optimal number of clusters for chronic mTBI symptoms is made in this example. The focus of this example is on predicting patterns in chronic symptoms. In order to measure how well symptom clusters can be predicted in an interpretable manner, the cluster labels were kept fixed in this example. The data indicate that 5 clusters is a reasonable place to start. [00216] For visualization purposes, the fuzzy nature of NMF was used to estimate uncertainty about cluster membership. To do so, each patient’s NMF coordinates was normalized by their sum, so they summed to one. Treating these normalized coordinates as a multinomial probability distribution over clusters, the entropy for each patient was computed, which is then taken as a measure of cluster uncertainty. [00217] The relationship between chronic symptom cluster and the following demographic and clinical variables was examined: age, gender, years of education, number of previous TBIs, and the severity of the worst previous TBI. The number and severity of previous TBIs were measured using the Ohio State University TBI Identification Method [48]. Separate DB2/ 47566363.1 47 Attorney Ref. No.: 118547-5008-WO linear models predicting age, education, and TBI number and severity were fit with cluster label, and a logistic regression predicting gender with cluster label. [00218] Baseline biological measurements. [00219] Data was for each patient from biological measures collected at baseline. Vital signs collected in emergency room (averaged within participants), lab results, radiologist-derived MRI and computed tomography findings, and S100B, NSE, and hsCRP candidate liquid biopsy-based biomarkers were included. Computed tomography findings included Marshall score, contusion, subarachnoid hemorrhage, skull fracture, epidural hematoma, inraventricular hemorrhage, subdural hematoma, extra-axial hematoma, and edema. Mean vitals, computed tomography scans, lab results, and blood biomarkers were collected immediately following hospitalization. MRI data were collected 14.8 ± 2.6 days (mean ± SD) after injury, but these features were eventually excluded due to being missing in too many patients. See Figure 3A for the full list of measurements considered in this example. [00220] All 1377 patients with 12 month symptom data had some subset of baseline biological measurements. However, because these biological measurements consisted of data from multiple modalities collected in different settings, there was much more missing data for these data than for the clinical scales collected at 12 months, described above. Figure 3 shows the percent missing for each baseline measurement we considered, before (top) and after (bottom) measures were dropped and patients missing from more than 50% of observations. A more lenient cut-off was used for biological data compared to clinical scales due to the large number of missing observations overall. Figure 3B shows the final list of features included after this cutoff. [00221] 1008 participants who met the cut-off for both chronic symptom and baseline biological data were left for use in the predictive models of this example. The remaining missing data were imputed, separately for training and test sets. Because the approach performed imputation separately for each random train/test split in the Monte Carlo procedure used to fit and evaluate the models, it is explained in more detail below. [00222] Symptom Prediction [00223] Overview [00224] To predict symptom cluster membership from baseline biological data, Elastic Net [49] generalized linear models (GLM) with a multinomial likelihood function implemented in the glmnet package in R [50] was used. Elastic Net models use a weighted combination of DB2/ 47566363.1 48 Attorney Ref. No.: 118547-5008-WO Ridge [51] (L2 regularization) and LASSO (L1 regularization) [52] penalties. A nested process of fitting, hyperparameter tuning, and validating models was utilized to avoid data leakage and estimate uncertainty about generalization performance. Specifically, Monte Carlo simulation was performed, repeatedly taking random samples of train/test splits and using 10-fold Cross Validation (CV) to optimize hyperparameters within a sample training split, before evaluating on the test set for that sample. The overall procedure for each sample was: [00225] Randomly sample train/test labels; [00226] Impute train and test sets separately with predictive mean matching; [00227] Optimize hyperparameters using 10-fold cross validation on training set; [00228] Evaluate AUC for each cluster on test set. [00229] This was repeated 100 times, providing a distribution of AUC scores and GLM coefficients for each cluster. [00230] Monte Carlo simulation and missing data imputation [00231] As noted above, an outer loop of Monte Carlo simulation was performed to estimate uncertainty about model coefficients and out of sample test performance, while marginalizing over optimized values of hyperparameters [53]. For each of 100 iterations, a binary train/test variable was randomly sampled for each patient from independent Bernoulli distributions with p=0.7, corresponding to an approximately 70/30 train/test split for each iteration. [00232] For each train/test split, the remaining missing baseline biological data was imputed using PMM. For imputation, character-coded and binary features were treated as factors, while other measurements were treated as numeric. NMF coordinates as well as cluster labels were included in imputation, which was performed separately for training and test sets in a given sample. After imputation, data were transformed into a matrix, with factors being dummy coded [54]. Columns of the training and test data matrices were then mean-centered and variance normalized. A column of ones was added to each of the matrices to represent an intercept term. [00233] Model optimization [00234] For each Monte Carlo iteration, model hyperparameters were optimized with 10-fold cross validation using the training set alone and then fit to the training set using the optimized DB2/ 47566363.1 49 Attorney Ref. No.: 118547-5008-WO hyperparameters. Specifically, Elastic Net regression represents a class of generalized linear models solving the following optimization problem: m ^^i^^n ^^^^ ( ^^^^, ^^^^, ^^^^ ) + ^^^^ [( 1 − ^^^^ ) ∥ ^^^^ ∥ 2 2/2 + ^^^^ ∥ ^^^^ ∥1 ]
Figure imgf000052_0001
where ^^^^( ^^^^, ^^^^, ^^^^) is a negative log-likelihood loss function combining response variables ^^^^, model coefficients ^^^^, and a feature matrix ^^^^. The form of this loss depends on the family of distributions assumed for the observations well as the link function (in this case, a multinomial likelihood with a softmax link). Note that in the multinomial case, both ^^^^ and ^^^^ are matrices, and ^^^^2 and ^^^^1 notation is used to refer to the Frobenius norm and the sum of ^^^^1 norms across classes, respectively. Weights were provided for each class (1 − # ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ to handle class imbalance.
Figure imgf000052_0002
[00235] Elastic Net hyperparameters: ^^^^ controls the overall strength of regularization, while ^^^^ controls the trade-off between ^^^^1 and ^^^^2 penalties. Elastic net reduces to LASSO when ^^^^ = 1 and to Ridge when ^^^^ = 0. [00236] A 10-fold cross validation was used within the training set to find optimal hyperparameters for each Monte Carlo iteration. A grid search was performed over values of ^^^^ and ^^^^, fitting the optimal ^^^^ coefficients for 10 cross validation folds while keeping the hyperparameters fixed to each ^^^^ and ^^^^ combination. Hyperparameters were selected that led to the lowest held out deviance averaged across cross validation folds, and then the ^^^^ parameters were refit to the whole training set with those hyperparameters. This process was repeated for each random train/test split. [00237] Model evaluation [00238] For each train/test split, the test set performance of the model fit to the training set with cross validation-optimized hyperparameters was evaluated. To do so, for each cluster the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) for the held out test data was measured, representing one versus all predictions for each cluster versus all others. This process was repeated 100 times to generate distributions of test set AUC scores for each cluster. The coefficients for each biomarker were also collected for each cluster, as well as the proportion of samples in which that coefficient was non-zero (because the Elastic Net model favored L1 regularization, which produced sparse coefficient matrices). DB2/ 47566363.1 50 Attorney Ref. No.: 118547-5008-WO [00239] To test whether these predictions were driven by between-cluster demographic differences, exploratory regression models were run across the combined training and test sets to include the demographic variables as confounds. After imputing missing biological data using the same predictive mean matching procedure described above, two logistic regressions were run (one for Cluster 5 versus all and one for Cluster 1 versus all) including all baseline biomarkers as well as age, gender, years of education, number of previous TBIs, and severity of the worst previous TBI. These models included no regularization term and were fit to all the data, with no training-test split. [00240] Results [00241] Scale correlations [00242] Individual item scores for psychiatric and daily living scales collected from patients who experienced an mTBI, 12 months after injury were aggregated. The Pearson correlation between scale items is shown in a heatmap Figure 4. The heatmap illustrates that most psychiatric scale items were correlated with each other, and that there were clear patterns of increased correlations both within- and between- items. For example, multiple clusters within the PCL-5 were apparent along the diagonal, while GOS-E and MPAI items were correlated with each other. Satisfaction with life was negatively correlated with other scales. [00243] Chronic symptom clustering [00244] NMF was used to cluster patients based on 12 month psychiatric symptoms. Results of fitting NMF with five dimensions are shown in Figure 5. The basis matrix W provides the loadings of each dimension/cluster onto the original clinical scales. Figure 5A shows a heatmap visualizing entries of W. The matrix is sparse, indicating that each factor is made up of a subset of scales. It is also apparent that the items were largely non-overlapping between factors, but that most factors did combine items across multiple scales. Factor 1 was made up of PTSD symptoms from the PCL-5, especially from re-experiencing and avoidance clusters. The top five items for this factor were all from these two clusters, with the highest three ranked items being “Upset by reminders”; “Repeated memories”; and “Avoiding memories, thoughts, or feelings”. Factor 2 was weighted most heavily for SWSL items, and thus could be considered an overall recovery cluster. Factor 3 combined depression symptoms from PHQ-9, BSI-18, and PCL-5 scales. Factor 4 reflected sleep symptoms and insomnia, combining items from PHQ-9, ISI, and PCL-5. Finally, Factor 5 reflected functional outcome and daily living problems, combining items from the MPAI4 and GOS-E scales. DB2/ 47566363.1 51 Attorney Ref. No.: 118547-5008-WO Each individual patient was assigned a cluster based on the dimension on which they scored highest in the coefficient matrix H. [00245] t-distributed Stochastic Neighbor Embedding (tSNE) [55] was used to visualize patient clusters (Figure 5B). Patient data points were identified by NMF cluster, and the opacity of each point was proportional to the negative entropy of the normalized coordinates (see Methods), illustrating uncertainty about cluster labels for each patient. Note that t-SNE was performed using the normalized Euclidean distance between patients in the space of the original item scores rather than in NMF space, to avoid introducing bias to the visualization due to the fact that NMF tends to produce sparse clustered coordinates by design. Some properties of the symptom clusters were apparent from the plot. While there is overlap between clusters, patients with high certainty tended to be separate from each other. It is also clear that most patients fell into the life satisfaction cluster, indicating that most experienced recovery by 12 months. Finally, the most distinct cluster was clearly the Functional Outcome cluster, indicating a small number of mTBI patients who were still experiencing trouble with daily living 12 months after injury. Figure 5C shows the number of participants assigned to each cluster. The most common cluster was the Life Satisfaction cluster (n = 581), while the lowest number of patients were assigned to the PTSD cluster (n = 87). [00246] Logistic regression was used to examine between cluster differences in demographic variables. Analysis of Variance (ANOVA) of model coefficients revealed that each variable examined varied significantly by cluster membership (Figure 6); (F_age(4, 1372) = 6.29, p < 0.0001; F_gender(4, 1372) = 2.93, p < 0.05; F_edu(4, 1352) = 14.7, p < 1e-11; F_(tbi,n)(4, 1372) = 8.11, p < 1e-05; F_(tbi,sev.)(4, 1372) = 9.13, p < 1e-06). Post-hoc pairwise comparisons with Tukey correction [56] revealed that the functional outcome cluster (Cluster 5) was significantly older than each of other clusters (p < 0.05 for all comparisons with Cluster 5). There were no other significant differences in age between pairs of clusters (p > 0.05 for all other comparisons). The life satisfaction cluster had the lowest proportion of females, however the only significant pairwise difference after posthoc correction was between life satisfaction and sleep disturbance clusters (p < 0.05). Years of education were significantly higher in the life satisfaction group than all other clusters (p < 0.01). Number and severity of prior TBIs were significantly higher in the depression cluster (p < 0.05). [00247] Predicting symptom clusters from baseline biology DB2/ 47566363.1 52 Attorney Ref. No.: 118547-5008-WO [00248] The primary goal of this example was to predict chronic symptom patterns from biological data collected at baseline after injury. Potential biomarker data collected from patients at baseline following injury were combined. This included vital signs, lab results, deriver computed tomography and MRI results, and three putative liquid biopsy-based biomarkers available on FITBIR: high sensitivity C-reactive protein (hs-CRP), S100 calcium binding protein B (S100B), and neuron-specific enolase (NSE). [00249] Multinomial Elastic Net models were used to predict the 12-month symptom clusters from baseline biological data, performing a nested cross-validation loop to optimize hyperparameters, and Monte Carlo sampling of train/test splits to evaluate predictions out of sample. Figure 7A shows the distribution of out-of-sample AUC scores for predicting each cluster versus all others across samples. The most accurate predictions were for the functional outcome deficit cluster (Cluster 5, AUC mean (SD) = 0.72 (0.042)). This indicates that prominent chronic functional outcome problems can be predicted with moderately high accuracy using data collected at the time of injury alone. There was also evidence that PTSD symptom cluster (Cluster 1) membership was predictable with moderate accuracy (mean AUC = 0.64, although there is much more uncertainty in this accuracy estimate compared to other clusters (SD = 0.084). Presumably this increased uncertainty was due in part to the small number of patients in the PTSD cluster. The other clusters (Life Satisfaction, Depression, and Sleep Disturbance) showed little evidence of predictability from the baseline data available for this study. [00250] To evaluate the features driving these predictions, the model coefficients for each class in each sample were extracted. Figures 7B, 7C, 7D, and 7E show the proportion of samples for which each feature was not equal to 0. Because the Elastic Net fit favored sparse feature spaces (by a high ^^^^ weighting ^^^^1 regularization) with most coefficients set exactly to 0, this proportion represents a measure of importance for each feature. Figures 7B, 7C, 7D and 7E also show the mean and standard deviation for the value of each feature’s coefficient across samples. For the functional outcome deficit cluster (Cluster 5), the most predictive baseline features were the presence of a subarachnoid hemorrhage, hsCRP levels, and the presence of an intraventricular hemorrhage, with all features contributing positively to cluster assignment. All three features were non-zero in >95% of models. For the PTSD cluster (Cluster 1), the most predictive features were the presence of a Subarachnoid Hemorrhage, mean sodium levels, and neuron-specific enolase level. All features were non-zero in >90% of models. While neuron-specific enolase and sodium contributed positively to PTSD cluster DB2/ 47566363.1 53 Attorney Ref. No.: 118547-5008-WO assignment, subarachnoid hemorrhage contributed negatively. Presumably, this was to distinguish mTBI patients with chronic prominent PTSD symptoms from those with chronic functional or daily living problems. Unsurprisingly given their lack of predictive accuracy, no features were clearly associated with Depression, Sleep Impairment, or Life Satisfaction clusters (Figure 8). [00251] Because the demographic variables examined in this example were all associated with chronic symptom cluster assignment (and some were associated with highly ranked biological features in our predictive models), analysis was conducted to determine whether these associations could not fully explain the biomarker prediction results. To do so, exploratory post-hoc regression models were run. For both PTSD and Functional Outcome clusters, no evidence that the results were driven by demographic-cluster associations was found. In fact, adjusting for baseline biomarker values, the only significant association between demographic variables and PTSD or Functional Outcome cluster assignment was a negative relationship between years of education and the PTSD cluster (p < 0.05). For the PTSD cluster (Cluster 1), Sodium, NSE, and subarachnoid hemorrhage were still significantly associated with cluster membership (all p < 0.005), in line with the predictive models. Similarly, the Functional Outcome cluster (Cluster 5), subarachnoid hemorrhage, intraventricular hemorrhage, and hsCRP remained significant (all p < 0.05). In addition, this analysis revealed a positive association between functional outcome and a computed tomography Marshall score above 4 (p < 0.001) as well as a negative association with platelet count (p < 0.001). Neither of these features were frequently selected in the predictive models of this example. See Table 1 and 2 below for full regression results. [00252] Table 1: Results of logistic regression predicting PTSD cluster from demographic and biological variables. Predictor Coefficient SE z p (Intercept) -7.61 24 -0.317 0.75089 HeartRate 0.00194 0.0143 0.136 0.89198 RespRate 0.00165 0.0774 0.0213 0.98298 Temp 0.0109 0.514 0.0212 0.9831 BldPressrSyst 0.012 0.0182 0.663 0.50742 BldPressrDiast -0.0171 0.0258 -0.662 0.50769 O2Satur -0.0545 0.132 -0.412 0.68042 ActPTT -0.012 0.0413 -0.291 0.77089 Calcium 0.052 0.363 0.143 0.88603 Creatinine 1.2 0.578 2.08 0.037975 Hematocrit 0.00103 0.0425 0.0241 0.98075 DB2/ 47566363.1 54 Attorney Ref. No.: 118547-5008-WO Predictor Coefficient SE z p InternationalNormalizedRatio 0.597 2.88 0.208 0.83543 Magnesium 1.19 0.849 1.4 0.16106 Neutrophil -0.000434 0.00799 -0.0544 0.95664 PlateletCount -0.00205 0.00335 -0.614 0.53932 Potassium 0.353 0.553 0.639 0.52301 ProthrombinTime -0.376 0.269 -1.4 0.16208 Sodium 0.0892 0.0597 1.49 0.13555 Urea 0.0107 0.0227 0.473 0.63636 WhiteBloodCellCount -0.0143 0.0622 -0.23 0.81773 CT_Marshall_234 -0.32 0.483 -0.662 0.50782 CT_Marshall_56 0.659 0.914 0.721 0.47108 CT_ContusionPresent 0.412 0.598 0.689 0.4909 CT_Subarachnoid_HemmorhagePresent -1.63 0.564 -2.89 0.0038442 CT_Skull_FractPresent 0.807 0.542 1.49 0.13646 CT_Epidural_HematomaPresent 0.0262 0.695 0.0377 0.96996 CT_Intraventricular_HemorrhagePresent -0.266 1.13 -0.235 0.814 CT_Subdural_Hematoma_SubAcutePresent -15.4 757 -0.0203 0.9838 CT_Extraaxial_HematomaPresent -0.27 0.895 -0.301 0.76306 CT_EdemaPresent -1.59 1.15 -1.37 0.16915 NSE 0.0112 0.00339 3.29 0.00099957 S100B -1.34 1.31 -1.02 0.30572 hsCRP -0.000501 0.00494 -0.101 0.91928 Age -0.0197 0.013 -1.52 0.12929 GenderMale -0.555 0.443 -1.25 0.21042 Edu -0.142 0.063 -2.26 0.023786 TBI_count -0.188 0.526 -0.358 0.7204 TBI_worst 0.0195 0.318 0.0612 0.95124 [00253] Table 2: Results of logistic regression predicting Functional Outcome cluster from demographic and biological variables Predictor Coefficient SE z p (Intercept) -19.6 14.2 -1.38 0.16825 HeartRate 0.002 0.00975 0.205 0.83767 RespRate 0.0313 0.0454 0.69 0.49014 Temp 0.436 0.314 1.39 0.16457 BldPressrSyst -0.0121 0.0112 -1.08 0.27962 BldPressrDiast 0.0148 0.0168 0.88 0.3786 O2Satur -0.0534 0.0704 -0.759 0.44795 ActPTT -0.00849 0.0229 -0.371 0.71077 Calcium 0.00388 0.25 0.0155 0.98762 Creatinine -0.339 0.455 -0.746 0.45547 Hematocrit -0.00505 0.0259 -0.195 0.84564 InternationalNormalizedRatio -5.89 1.84 -3.2 0.001393 Magnesium -1.12 0.556 -2.02 0.043784 Neutrophil 0.000456 0.0056 0.0815 0.93505 DB2/ 47566363.1 55 Attorney Ref. No.: 118547-5008-WO Predictor Coefficient SE z p PlateletCount 0.000178 0.00221 0.0805 0.93584 Potassium -0.0249 0.373 -0.0667 0.94685 ProthrombinTime 0.783 0.217 3.61 0.00030524 Sodium 0.025 0.0374 0.668 0.50395 Urea 0.0211 0.0141 1.5 0.13253 WhiteBloodCellCount 0.00489 0.0388 0.126 0.89982 CT_Marshall_234 0.499 0.334 1.5 0.1347 CT_Marshall_56 1.72 0.51 3.37 0.00074582 CT_ContusionPresent -0.217 0.312 -0.697 0.48601 CT_Subarachnoid_HemmorhagePresent 0.65 0.305 2.13 0.033156 CT_Skull_FractPresent -0.391 0.313 -1.25 0.21144 CT_Epidural_HematomaPresent -0.912 0.475 -1.92 0.054856 CT_Intraventricular_HemorrhagePresent 0.837 0.381 2.2 0.027809 CT_Subdural_Hematoma_SubAcutePresent -0.76 0.767 -0.99 0.3221 CT_Extraaxial_HematomaPresent 0.709 0.476 1.49 0.13612 CT_EdemaPresent -0.0766 0.375 -0.204 0.83801 NSE 4.9e-05 0.00291 0.0168 0.98657 S100B 1.35 0.675 2 0.045269 hsCRP 0.0087 0.00241 3.61 0.00030936 Age 0.0197 0.0077 2.56 0.010616 GenderMale -0.209 0.273 -0.764 0.44516 Edu -0.0644 0.0391 -1.65 0.099659 TBI_count -0.164 0.375 -0.437 0.66221 TBI_worst 0.0869 0.231 0.376 0.70698 [00254] These results were consistent with the baseline biomarkers providing unique information about patient outcomes beyond available demographic data. [00255] Discussion. [00256] To summarize, the heterogeneity inherent to TBI could be meaningfully parsed into biologically relevant clusters at the chronic stage. Specifically, functional outcome deficits could be accurately predicted using baseline hsCRP and CT measurements. Furthermore, a cluster characterized by PTSD re-experiencing and avoidance symptoms was found, which could be predicted from neuron-specific enolase and sodium level features. [00257] These findings align with previous reports, underscoring hsCRP [27] and subarachnoid and intraventricular hemorrhage CT findings [25] as accurate and reliable predictors of long-term functional outcomes. These findings add to previous understanding by combining baseline measurements from multiple modalities, and employing machine learning to directly predict outcomes at the individual level while evaluating model performance using out-of-sample predictions. This approach extends previous hsCRP DB2/ 47566363.1 56 Attorney Ref. No.: 118547-5008-WO findings to 12 months after injury, and explores the feasibility of predicting recovery beyond functional outcome deficits. [00258] These results provide direct support to the idea that baseline blood and brain imaging measures hold promise as prognostic biomarkers following brain injury. The development of biomarkers for prognostic contexts of use remains a major goal in precision neurology. [00259] In conclusion, our study not only highlights the potential of using baseline biomarkers and imaging data for prognostic purposes but also sets the stage for future research aimed at refining these predictive models. By advancing our ability to forecast long- term outcomes in mTBI patients, we move closer to realizing the goal of personalized neurotrauma care, ultimately improving the quality of life for individuals affected by brain injury. [00260] References [00261] 1. Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic brain injury–related emergency department visits, hospitalizations, and deaths—united states, 2007 and 2013. MMWR Surveillance Summaries.2017;66:1. [00262] 2. Foks KA, Cnossen MC, Dippel DW, Maas AI, Menon D, Naalt J van der, et al. Management of mild traumatic brain injury at the emergency department and hospital admission in europe: A survey of 71 neurotrauma centers participating in the CENTER-TBI study. Journal of Neurotrauma.2017;34:2529–2535. [00263] 3. Levin HS, Diaz-Arrastia RR. Diagnosis, prognosis, and clinical management of mild traumatic brain injury. The Lancet Neurology.2015;14:506–517. [00264] 4. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: A brief overview. The Journal of Head Trauma Rehabilitation. 2006;21:375–378. [00265] 5. Mac Donald CL, Barber J, Jordan M, Johnson AM, Dikmen S, Fann JR, et al. Early clinical predictors of 5-year outcome after concussive blast traumatic brain injury. JAMA Neurology.2017;74:821–829. [00266] 6. Pattinson CL, Shahim P, Taylor P, Dunbar K, Guedes V, Motamedi V, et al. Elevated tau in military personnel relates to chronic symptoms following traumatic brain injury. The Journal of Head Trauma Rehabilitation.2020;35:66. DB2/ 47566363.1 57 Attorney Ref. No.: 118547-5008-WO [00267] 7. Radhakrishnan R, Garakani A, Gross LS, Goin MK, Pine J, Slaby AE, et al. Neuropsychiatric aspects of concussion. The Lancet Psychiatry.2016;3:1166–1175. [00268] 8. McCrea M, Guskiewicz KM, Marshall SW, Barr W, Randolph C, Cantu RC, et al. Acute effects and recovery time following concussion in collegiate football players: The NCAA concussion study. Jama.2003;290:2556–2563. [00269] 9. Nelson LD, Furger RE, Ranson J, Tarima S, Hammeke TA, Randolph C, et al. Acute clinical predictors of symptom recovery in emergency department patients with uncomplicated mild traumatic brain injury or non-traumatic brain injuries. Journal of Neurotrauma.2018;35:249–259. [00270] 10. Rohling ML, Binder LM, Demakis GJ, Larrabee GJ, Ploetz DM, Langhinrichsen-Rohling J. A meta-analysis of neuropsychological outcome after mild traumatic brain injury: Re-analyses and reconsiderations of binder et al., frencham et al., and pertab et al. The Clinical Neuropsychologist.2011;25:608–623. [00271] 11. Rogers JM, Read CA. Psychiatric comorbidity following traumatic brain injury. Brain Injury.2007;21:1321–1333. [00272] 12. Hammond FM, Corrigan JD, Ketchum JM, Malec JF, Dams-O’Conner K, Hart T, et al. Prevalence of medical and psychiatric comorbidities following traumatic brain injury. The Journal of Head Trauma Rehabilitation.2019;34:E1. [00273] 13. Ponsford J, Cameron P, Fitzgerald M, Grant M, Mikocka-Walus A, Schönberger M. Predictors of postconcussive symptoms 3 months after mild traumatic brain injury. Neuropsychology.2012;26:304. [00274] 14. Alway Y, McKay A, Gould KR, Johnston L, Ponsford J. Factors associated with posttraumatic stress disorder following moderate to severe traumatic brain injury: A prospective study. Depression and Anxiety.2016;33:19–26. [00275] 15. Tanev KS, Pentel KZ, Kredlow MA, Charney ME. PTSD and TBI co-morbidity: Scope, clinical presentation and treatment options. Brain Injury.2014;28:261–270. [00276] 16. Hajek CA, Yeates KO, Gerry Taylor H, Bangert B, Dietrich A, Nuss KE, et al. Relationships among post-concussive symptoms and symptoms of PTSD in children following mild traumatic brain injury. Brain Injury.2010;24:100–109. [00277] 17. Cnossen MC, Scholten AC, Lingsma HF, Synnot A, Haagsma J, Steyerberg PEW, et al. Predictors of major depression and posttraumatic stress disorder following DB2/ 47566363.1 58 Attorney Ref. No.: 118547-5008-WO traumatic brain injury: A systematic review and meta-analysis. The Journal of Neuropsychiatry and Clinical Neurosciences.2017;29:206–224. [00278] 18. Fann JR, Uomoto JM, Katon WJ. Cognitive improvement with treatment of depression following mild traumatic brain injury. Psychosomatics.2001;42:48–54. [00279] 19. Rapoport MJ, Mccullagh S, Streiner D, Feinstein A. The clinical significance of major depression following mild traumatic brain injury. Psychosomatics.2003;44:31–37. [00280] 20. Hellewell SC, Beaton CS, Welton T, Grieve SM. Characterizing the risk of depression following mild traumatic brain injury: A meta-analysis of the literature comparing chronic mTBI to non-mTBI populations. Frontiers in Neurology.2020;11:350. [00281] 21. Mallya S, Sutherland J, Pongracic S, Mainland B, Ornstein TJ. The manifestation of anxiety disorders after traumatic brain injury: A review. Journal of Neurotrauma.2015;32:411–421. [00282] 22. Moore EL, Terryberry-Spohr L, Hope DA. Mild traumatic brain injury and anxiety sequelae: A review of the literature. Brain Injury.2006;20:117–132. [00283] 23. Hart T, Fann JR, Chervoneva I, Juengst SB, Rosenthal JA, Krellman JW, et al. Prevalence, risk factors, and correlates of anxiety at 1 year after moderate to severe traumatic brain injury. Archives of Physical Medicine and Rehabilitation.2016;97:701–707. [00284] 24. Hoge CW, Castro CA. Treatment of generalized war-related health concerns: Placing TBI and PTSD in context. Jama.2014;312:1685–1686. [00285] 25. Yuh EL, Jain S, Sun X, Pisică D, Harris MH, Taylor SR, et al. Pathological computed tomography features associated with adverse outcomes after mild traumatic brain injury: A TRACK-TBI study with external validation in CENTER-TBI. JAMA Neurology. 2021;78:1137–1148. [00286] 26. Kulbe JR, Jain S, Nelson LD, Korley FK, Mukherjee P, Sun X, et al. Association of day-of-injury plasma glial fibrillary acidic protein concentration and six- month posttraumatic stress disorder in patients with mild traumatic brain injury. Neuropsychopharmacology.2022;47:2300–2308. [00287] 27. Xu LB, Yue JK, Korley F, Puccio AM, Yuh EL, Sun X, et al. High-sensitivity c- reactive protein is a prognostic biomarker of six-month disability after traumatic brain injury: Results from the TRACK-TBI study. Journal of Neurotrauma.2021;38:918–927. DB2/ 47566363.1 59 Attorney Ref. No.: 118547-5008-WO [00288] 28. Michopoulos V, Rothbaum AO, Jovanovic T, Almli LM, Bradley B, Rothbaum BO, et al. Association of CRP genetic variation and CRP level with elevated PTSD symptoms and physiological responses in a civilian population with high levels of trauma. American Journal of Psychiatry.2015;172:353–362. [00289] 29. Nelson LD, Temkin NR, Dikmen S, Barber J, Giacino JT, Yuh E, et al. Recovery after mild traumatic brain injury in patients presenting to US level i trauma centers: A transforming research and clinical knowledge in traumatic brain injury (TRACK-TBI) study. JAMA Neurology.2019;76:1049–1059. [00290] 30. Thompson HJ, Vavilala MS, Rivara FP. Common data elements and federal interagency traumatic brain injury research informatics system for TBI research. Annu Rev Nurs Res.2015;33:1–11. [00291] 31. Teasdale G, Maas A, Lecky F, Manley G, Stocchetti N, Murray G. The glasgow coma scale at 40 years: Standing the test of time. The Lancet Neurology.2014;13:844–854. [00292] 32. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine.2001;16:606–613. [00293] 33. McMillan T, Wilson L, Ponsford J, Levin H, Teasdale G, Bond M. The glasgow outcome scale—40 years of application and refinement. Nature Reviews Neurology. 2016;12:477–485. [00294] 34. Bastien CH, Vallières A, Morin CM. Validation of the insomnia severity index as an outcome measure for insomnia research. Sleep Medicine.2001;2:297–307. [00295] 35. Pavot W, Diener E. The satisfaction with life scale and the emerging construct of life satisfaction. The Journal of Positive Psychology.2008;3:137–152. [00296] 36. Bellon K, Malec JF, Kolakowsky-Hayner SA. Mayo-portland adaptability inventory-4. The Journal of Head Trauma Rehabilitation.2012;27:314–316. [00297] 37. Derogatis LR, Fitzpatrick M. The SCL-90-r, the brief symptom inventory (BSI), and the BSI-18.2004.2004. [00298] 38. Vink G, Frank LE, Pannekoek J, Van Buuren S. Predictive mean matching imputation of semicontinuous variables. Statistica Neerlandica.2014;68:61–90. [00299] 39. Zhang Z. Multiple imputation with multivariate imputation by chained equation (MICE) package. Annals of Translational Medicine.2016;4. DB2/ 47566363.1 60 Attorney Ref. No.: 118547-5008-WO [00300] 40. Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization. Nature.1999;401:788–791. [00301] 41. Wang Y-X, Zhang Y-J. Nonnegative matrix factorization: A comprehensive review. IEEE Transactions on Knowledge and Data Engineering.2012;25:1336–1353. [00302] 42. Gaujoux R, Seoighe C. A flexible r package for nonnegative matrix factorization. BMC Bioinformatics.2010;11:1–9. [00303] 43. Binesh N, Rezghi M. Fuzzy clustering in community detection based on nonnegative matrix factorization with two novel evaluation criteria. Applied Soft Computing. 2018;69:689–703. [00304] 44. Lazar C, Doncescu A. Non negative matrix factorization clustering capabilities; application on multivariate image segmentation.2009 international conference on complex, intelligent and software intensive systems, IEEE; 2009. p.924–929. [00305] 45. Zhang S, Wang R-S, Zhang X-S. Uncovering fuzzy community structure in complex networks. Physical Review E.2007;76:046103. [00306] 46. Brunet J-P, Tamayo P, Golub TR, Mesirov JP. Metagenes and molecular pattern discovery using matrix factorization. Proceedings of the National Academy of Sciences. 2004;101:4164–4169. [00307] 47. Shahapure KR, Nicholas C. Cluster quality analysis using silhouette score.2020 IEEE 7th international conference on data science and advanced analytics (DSAA), IEEE; 2020. p.747–748. [00308] 48. Corrigan JD, Bogner J. Initial reliability and validity of the ohio state university TBI identification method. The Journal of Head Trauma Rehabilitation.2007;22:318–329. [00309] 49. Zou H, Hastie T. Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society: Series B (Statistical Methodology).2005;67:301– 320. [00310] 50. Hastie T, Qian J, Tay K. An introduction to glmnet. CRAN R Repositary.2021. 2021. [00311] 51. Hoerl AE, Kennard RW. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics.1970;12:55–67. DB2/ 47566363.1 61 Attorney Ref. No.: 118547-5008-WO [00312] 52. Tibshirani R. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological).1996;58:267–288. [00313] 53. Hong F, Tian L, Devanarayan V. Improving the robustness of variable selection and predictive performance of regularized generalized linear models and cox proportional hazard models. Mathematics.2023;11:557. [00314] 54. Berry KJ, Mielke Jr PW, Iyer HK. Factorial designs and dummy coding. Perceptual and Motor Skills.1998;87:919–927. [00315] 55. Van der Maaten L, Hinton G. Visualizing data using t-SNE. Journal of Machine Learning Research.2008;9. [00316] 56. Abdi H, Williams LJ. Tukey’s honestly significant difference (HSD) test. Encyclopedia of Research Design.2010;3:1–5. REFERENCES CITED AND ALTERNATIVE EMBODIMENTS [00317] All references cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual publication or patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety for all purposes. [00318] The present invention can be implemented as a computer program product that includes a computer program mechanism embedded in a non-transitory computer-readable storage medium. For instance, the computer program product could contain instructions for operating the user interfaces disclosed herein and described with respect to the Figures. These program modules can be stored on a CD-ROM, DVD, magnetic disk storage product, USB key, or any other non-transitory computer readable data or program storage product. [00319] Many modifications and variations of this invention can be made without departing from its spirit and scope, as will be apparent to those skilled in the art. The specific embodiments described herein are offered by way of example only. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. The invention is to be limited only by the terms of the appended claims, along with the full scope of equivalents to which such claims are entitled. DB2/ 47566363.1 62

Claims

Attorney Ref. No.: 118547-5008-WO WHAT IS CLAIMED IS: 1. A method for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit, the method comprising: obtaining one or more liquid biopsy-based baseline biomarker results of the mTBI subject within a month of incurring a traumatic brain injury, wherein the one or more liquid biopsy-based baseline biomarker results comprises a baseline high-sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject; obtaining a radiographic modality finding of the mTBI subject within a month of incurring the traumatic brain injury; and providing an indication, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit. 2. The method of claim 1, wherein the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit. 3. The method of claim 1, wherein the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit. 4. The method of any one of claims 1-3, wherein the radiographic modality finding consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage. 5. The method of any one of claims 1-3, wherein the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. 6. The method of any one of claims 1-3, wherein the radiographic modality finding of the mTBI subject comprises (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. DB2/ 47566363.1 63 Attorney Ref. No.: 118547-5008-WO 7. The method of any one of claims 1-3, wherein the radiographic modality finding of the mTBI subject comprises: a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture. 8. The method of any one of claims 1-7, the method further comprising obtaining a baseline heart rate, a diastolic blood pressure, a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject, and wherein the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline heart rate, the diastolic blood pressure, the respiratory rate, the temperature, the systolic blood pressure, or any combination thereof, of the mTBI subject. 9. The method of any one of claims 1-8, wherein the one or more liquid biopsy-based baseline biomarker results comprises a neuron-specific enolase (NSE) measurement or a S100B measurement of the mTBI subject. 10. The method of any one of claims 1-9, the method further comprising: obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject, and the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject. 11. The method of any one of claims 1-10, wherein the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15. 12. The method of any one of claims 1-11, wherein the providing the indication comprises: DB2/ 47566363.1 64 Attorney Ref. No.: 118547-5008-WO inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy- based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit. 13. The method of claim 12, wherein the model is an Elastic Net model with a multinomial likelihood function. 14. The method of claim 12, wherein the model is a nonnegative matrix factorization model. 15. The method of claim 12, wherein the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof. 16. The method of any one of claims 12-15, wherein the inputting further comprises inputting one or more covariates into the model. 17. The method of claim 16, wherein the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof. 18. The method of any one of claims 1-17, wherein the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises treating the mTBI subject based on the indication. DB2/ 47566363.1 65 Attorney Ref. No.: 118547-5008-WO 19. The method of claim 18, wherein the treating the mTBI subject based on the indication comprises administering psychotherapy to the mTBI subject. 20. The method of claim 18 or 19, wherein the treating the mTBI subject based on the indication comprises administering physical therapy to the mTBI subject. 21. The method of any one of claims 18-20, wherein the treating the mTBI subject based on the indication comprises administering speech therapy to the mTBI subject. 22. The method of any one of claims 18-21, wherein the treating the mTBI subject based on the indication comprises administering occupational therapy to the mTBI subject. 23. The method of any one of claims 18-22, wherein the treating the mTBI subject based on the indication comprises administering melatonin or zolipidem to the mTBI subject. 24. The method of any one of claims 18-23, wherein the treating the mTBI subject based on the indication comprises administering a selective serotonin reuptake inhibitor to the mTBI subject. 25. The method of any one of claims 18-24, wherein the treating the mTBI subject based on the indication comprises administering methylphenidate or amphetamine salt to the mTBI subject. 26. The method of any one of claims 18-25, wherein the treating the mTBI subject based on the indication comprises administering cyclobenzaprine to the mTBI subject. 27. The method of any one of claims 1-17, wherein the indication is that the mTBI subject will incur the chronic functional outcome deficit, and the method further comprises adjusting a treatment regime of the mTBI subject based on the indication. 28. The method of claim 27, wherein the adjusting the treatment regime comprises increasing a frequency or duration of psychotherapy for the mTBI subject. DB2/ 47566363.1 66 Attorney Ref. No.: 118547-5008-WO 29. The method of claim 27 or 28, wherein the adjusting the treatment regime comprises increasing a frequency or a duration of physical therapy for the mTBI subject. 30. The method of any one of claims 27-29, wherein the adjusting the treatment regime comprises increasing a frequency or a duration of speech therapy for the mTBI subject. 31. The method of any one of claims 27-30, wherein the adjusting the treatment regime comprises increasing a frequency or a duration of occupational therapy for the mTBI subject. 32. The method of any one of claims 27-31, wherein the adjusting the treatment regime comprises increasing a dosage of melatonin or zolipidem given to the mTBI subject or increasing a length of time the melatonin or zolipidem is given to the mTBI subject. 33. The method of any one of claims 27-32, wherein the adjusting the treatment regime comprises increasing a dosage of a selective serotonin reuptake inhibitor given to the mTBI subject or increasing a length of time the selective serotonin reuptake inhibitor is given to the mTBI subject. 34. The method of any one of claims 27-36, wherein the adjusting the treatment regime comprises increasing a dosage of methylphenidate or amphetamine salt given to the mTBI subject or increasing a length of time the methylphenidate or amphetamine salt is given to the mTBI subject 35. The method of any one of claims 27-36, wherein the adjusting the treatment regime comprises increasing a dosage of cyclobenzaprine give to the mTBI subject or increasing a length of time the cyclobenzaprine is given to the mTBI subject. 36. The method of any one of claims 1-35, wherein the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury. DB2/ 47566363.1 67 Attorney Ref. No.: 118547-5008-WO 37. The method of any one of claims 1-35, wherein the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury. 38. The method of any one of claims 1-37, wherein the radiographic modality finding is obtained through magnetic resonance imaging. 39. The method of any one of claims 1-37, wherein the radiographic modality finding is obtained through computed tomography. 40. The method of any one of claims 1-39, wherein the chronic functional outcome deficit is a Glasgow Outcome Scale-Extended (GOS-E) score of less than or equal to 7, twelve months after incurring the traumatic brain injury. 41. The method of any one of claims 1-39, wherein the chronic functional outcome deficit is a GOS-E score of less than or equal to 6, twelve months after incurring the traumatic brain injury. 42. A computer system for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit, the computer system comprising: one or more processors; memory storing one or more programs to be executed by the one or more processors; the one or more programs comprising instructions for: obtaining one or more liquid biopsy-based baseline biomarker results of the mTBI subject within a month of incurring a traumatic brain injury, wherein the one or more liquid biopsy-based baseline biomarker results comprises a baseline high-sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject; obtaining a radiographic modality finding of the mTBI subject within a month of incurring the traumatic brain injury; and providing an indication, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic DB2/ 47566363.1 68 Attorney Ref. No.: 118547-5008-WO modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit. 43. The computer system of claim 42, wherein the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit. 44. The computer system of claim 42, wherein the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit. 45. The computer system of any one of claims 42-44, wherein the radiographic modality finding consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage. 46. The computer system of any one of claims 42-44, wherein the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. 47. The computer system of any one of claims 42-44, wherein the radiographic modality finding of the mTBI subject comprise (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. 48. The computer system of any one of claims 42-44, wherein the radiographic modality finding of the mTBI subject comprises: a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture. 49. The computer system of any one of claims 42-48, the one or more programs further comprising instructions for further obtaining a baseline heart rate, a diastolic blood pressure, DB2/ 47566363.1 69 Attorney Ref. No.: 118547-5008-WO a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject, and wherein the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline heart rate, the diastolic blood pressure, the respiratory rate, the temperature, the systolic blood pressure, or any combination thereof, of the mTBI subject. 50. The computer system of any one of claims 42-48, wherein the one or more liquid biopsy- based baseline biomarker results comprises a neuron-specific enolase (NSE) measurement or a S100B measurement of the mTBI subject. 51. The computer system of any one of claims 42-50, wherein the one or more programs further comprise instructions for: obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject, and the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject. 52. The computer system of any one of claims 42-51, wherein the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15. 53. The computer system of any one of claims 42-52, wherein the providing the indication comprises: inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy- based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit. DB2/ 47566363.1 70 Attorney Ref. No.: 118547-5008-WO 54. The computer system of claim 53, wherein the model is an Elastic Net model with a multinomial likelihood function. 55. The computer system of claim 53, wherein the model is a nonnegative matrix factorization model. 56. The computer system of claim 53, wherein the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof. 57. The computer system of any one of claims 53-56, wherein the inputting further comprises inputting one or more covariates into the model. 58. The computer system of claim 57, wherein the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof. 59. The computer system of any one of claims 42-58, wherein the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury. 60. The computer system of any one of claims 42-58, wherein the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and DB2/ 47566363.1 71 Attorney Ref. No.: 118547-5008-WO the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury. 61. The computer system of any one of claims 42-60, wherein the radiographic modality finding is obtained through magnetic resonance imaging. 62. The computer system of any one of claims 42-60, wherein the radiographic modality finding is obtained through computed tomography. 63. The computer system of any one of claims 42-62, wherein the chronic functional outcome deficit is a Glasgow Outcome Scale-Extended (GOS-E) score of less than or equal to 7, twelve months after incurring the traumatic brain injury. 64. The computer system of any one of claims 42-62, wherein the chronic functional outcome deficit is a GOS-E score of less than or equal to 6, twelve months after incurring the traumatic brain injury. 65. A non-transitory computer readable storage medium storing one or more programs for determining whether a mild traumatic brain injury (mTBI) subject will incur a chronic functional outcome deficit, the one or more programs configured for execution by a computer, the one or more programs comprising instructions for: obtaining one or more liquid biopsy-based baseline biomarker results of the mTBI subject within a month of incurring a traumatic brain injury, wherein the one or more liquid biopsy-based baseline biomarker results comprises a baseline high-sensitivity C-reactive protein (hsCRP) measurement of the mTBI subject; obtaining a radiographic modality finding of the mTBI subject within a month of incurring the traumatic brain injury; and providing an indication, based at least on a combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, as to whether the mTBI subject will incur the chronic functional outcome deficit. DB2/ 47566363.1 72 Attorney Ref. No.: 118547-5008-WO 66. The non-transitory computer readable storage medium of claim 65, wherein the indication is in the form of a binary indication as to whether the mTBI subject will incur chronic functional outcome deficit. 67. The non-transitory computer readable storage medium of claim 65, wherein the indication is in the form of a probability or likelihood that the mTBI subject will incur chronic functional outcome deficit. 68. The non-transitory computer readable storage medium of any one of claims 65-67, wherein the radiographic modality finding consists of an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage. 69. The non-transitory computer readable storage medium of any one of claims 65-67, wherein the radiographic modality finding of the mTBI subject consists of an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. 70. The non-transitory computer readable storage medium of any one of claims 65-67, wherein the radiographic modality finding of the mTBI subject comprises (a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage and (b) an indication as to whether or not the mTBI subject has an intraventricular hemorrhage. 71. The non-transitory computer readable storage medium of any one of claims 65-67, wherein the radiographic modality finding of the mTBI subject comprises: a) an indication as to whether or not the mTBI subject has a subarachnoid hemorrhage or an indication as to whether or not the mTBI subject has an intraventricular hemorrhage, and b) an indication as to whether or not the mTBI subject has an epidural hematoma, a Marshall 56, a Marshall 234, an edema, a contusion, a subacute subdural hematoma, an extraaxial hematoma, or skull fracture. 72. The non-transitory computer readable storage medium of any one of claims 65-71, wherein the one or more programs further comprise instructions for obtaining a baseline heart rate, a diastolic blood pressure, a respiratory rate, a temperature, a systolic blood pressure, or any combination thereof, of the mTBI subject, and wherein DB2/ 47566363.1 73 Attorney Ref. No.: 118547-5008-WO the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline heart rate, the diastolic blood pressure, the respiratory rate, the temperature, the systolic blood pressure, or any combination thereof, of the mTBI subject. 73. The non-transitory computer readable storage medium of any one of claims 65-72, wherein the one or more liquid biopsy-based baseline biomarker results comprises a neuron- specific enolase (NSE) measurement or a S100B measurement of the mTBI subject. 74. The non-transitory computer readable storage medium of any one of claims 65-73, wherein the one or more programs further comprise instructions for: obtaining a baseline potassium level, a baseline sodium level, a baseline platelet count level, a baseline hematocrit level, a baseline blood O2 saturation level, a baseline white blood cell count, or any combination thereof, of the mTBI subject, and the indication as to whether the mTBI subject will incur the chronic functional outcome deficit is further based on the baseline potassium level, the baseline sodium level, the baseline platelet count level, the baseline hematocrit level, the baseline blood O2 saturation level, the baseline white blood cell count, or any combination thereof, of the mTBI subject. 75. The non-transitory computer readable storage medium of any one of claims 65-74, wherein the mTBI subject has a baseline Glasgow Coma Scale (GCS) score in the range of 13-15. 76. The non-transitory computer readable storage medium of any one of claims 65-74, wherein the providing the indication comprises: inputting at least the combination of (i) the one or more liquid biopsy-based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject, into a model; and responsive to inputting at least on the combination of (i) the one or more liquid biopsy- based baseline biomarker results of the mTBI subject and (ii) the radiographic modality finding of the mTBI subject into the model, receiving as output from the model the indication as to whether the mTBI subject will incur the chronic functional outcome deficit. DB2/ 47566363.1 74 Attorney Ref. No.: 118547-5008-WO 77. The non-transitory computer readable storage medium of claim 76, wherein the model is an Elastic Net model with a multinomial likelihood function. 78. The non-transitory computer readable storage medium of claim 76, wherein the model is a nonnegative matrix factorization model. 79. The non-transitory computer readable storage medium of claim 76, wherein the model is a clustering model, a logistic regression model, a neural network, a support vector machine, a Naive Bayes model, a nearest neighbors model, a random forest model, a decision tree model, a boosted trees model, a multinomial logistic regression model, a linear model, a linear regression model, a Gradient Boosting model, a mixture model, a hidden Markov model, a Gaussian NB model, a linear discriminant analysis model, or any combination or ensemble or boosted ensemble thereof. 80. The non-transitory computer readable storage medium of any one of claims 77-79, wherein the inputting further comprises inputting one or more covariates into the model. 81. The non-transitory computer readable storage medium of claim 80, wherein the one or more covariates is an age of the mTBI subject, a sex of the mTBI subject, a number of years of education of the mTBI subject, a number of prior traumatic brain injuries incurred by the mTBI subject, or a severity of a worst prior traumatic brain injury incurred by the mTBI subject, or any combination thereof. 82. The non-transitory computer readable storage medium of any one of claims 65-81, wherein the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within a week of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within a week of incurring the traumatic brain injury. 83. The non-transitory computer readable storage medium of any one of claims 65-81, wherein DB2/ 47566363.1 75 Attorney Ref. No.: 118547-5008-WO the one or more liquid biopsy-based baseline biomarker results of the mTBI subject are obtained within two days of incurring the traumatic brain injury, and the radiographic modality finding of the mTBI subject is obtained within two days of incurring the traumatic brain injury. 84. The non-transitory computer readable storage medium of any one of claims 65-83, wherein the radiographic modality finding is obtained through magnetic resonance imaging. 85. The non-transitory computer readable storage medium of any one of claims 65-83, wherein the radiographic modality finding is obtained through computed tomography. 86. The non-transitory computer readable storage medium of any one of claims 65-85, wherein the chronic functional outcome deficit is a Glasgow Outcome Scale-Extended (GOS- E) score of less than or equal to 7, twelve months after incurring the traumatic brain injury. 87. The non-transitory computer readable storage medium of any one of claims 65-85, wherein the chronic functional outcome deficit is a GOS-E score of less than or equal to 6, twelve months after incurring the traumatic brain injury. DB2/ 47566363.1 76
PCT/US2024/018646 2023-03-06 2024-03-06 Systems, methods, and devices for using prognostic indicators of chronic outcome in traumatic brain injury Pending WO2024186885A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US202363450325P 2023-03-06 2023-03-06
US63/450,325 2023-03-06
US202363472084P 2023-06-09 2023-06-09
US63/472,084 2023-06-09

Publications (1)

Publication Number Publication Date
WO2024186885A1 true WO2024186885A1 (en) 2024-09-12

Family

ID=90731376

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2024/018646 Pending WO2024186885A1 (en) 2023-03-06 2024-03-06 Systems, methods, and devices for using prognostic indicators of chronic outcome in traumatic brain injury

Country Status (1)

Country Link
WO (1) WO2024186885A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050130230A1 (en) * 2003-09-23 2005-06-16 Antoni Davalos Cellular fibronectin as a diagnostic marker in stroke and methods of use thereof
WO2018217792A1 (en) * 2017-05-23 2018-11-29 Immunarray USA, Inc. Biomarker levels and neuroimaging for detecting, monitoring and treating brain injury or trauma
WO2020231846A1 (en) * 2019-05-10 2020-11-19 Abbott Laboratories Evaluating biomarkers along with advanced magnetic resonance imaging procedures in a human subject that has sustained or may have sustained a head injury

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050130230A1 (en) * 2003-09-23 2005-06-16 Antoni Davalos Cellular fibronectin as a diagnostic marker in stroke and methods of use thereof
WO2018217792A1 (en) * 2017-05-23 2018-11-29 Immunarray USA, Inc. Biomarker levels and neuroimaging for detecting, monitoring and treating brain injury or trauma
WO2020231846A1 (en) * 2019-05-10 2020-11-19 Abbott Laboratories Evaluating biomarkers along with advanced magnetic resonance imaging procedures in a human subject that has sustained or may have sustained a head injury

Non-Patent Citations (117)

* Cited by examiner, † Cited by third party
Title
"Neuropsychological rehabilitation of mild traumatic brain injury", BRAIN INJURY, vol. 10, no. 4, 2009, pages 277 - 286
ABDI HWILLIAMS LJ: "Encyclopedia of Research Design", vol. 3, 2010, article "Tukey's honestly significant difference (HSD) test", pages: 1 - 5
ALARIE ET AL.: "Physical activity interventions in rehabilitation programs for outpatients with mild traumatic brain injury", RESEARCH QUARTERLY FOR EXERCISE AND SPORT, vol. 93, no. 4, 2022, pages 851 - 860
ALASHRAM: "Optimizing gait ability after task-oriented circuit class training in posttraumatic brain injury: a case report", INDIAN J PHYS MED REHABIL., vol. 30, no. 4, 2019, pages 112 - 116
ALEXANDER: "Diffusion tensor imaging of the brain, ", NEUROTHERAPEUTICS, vol. 4, 2007, pages 316 - 329, XP055916523, DOI: 10.1016/j.nurt.2007.05.011
ALLEN: "Neurosensory disorders in mild traumatic brain injury", 2019, ACADEMIC PRESS, article "Cognitive rehabilitation for mild traumatic brain injury (mTBI", pages: 357 - 379
ALWAGEED: "Detection of Subarachnoid Hemorrhage in Computed Tomography Using Association Rules Mining", COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE, vol. 2022, no. 1133819, 2022
ALWAY YMCKAY AGOULD KRJOHNSTON LPONSFORD J.: "Factors associated with posttraumatic stress disorder following moderate to severe traumatic brain injury: A prospective study", DEPRESSION AND ANXIETY, vol. 33, 2016, pages 19 - 26, XP072006316, DOI: 10.1002/da.22396
BAINSABDIJADID: "StatPearls", 10 April 2023, STATPEARLS PUBLISHING, article "Major Depressive Disorder"
BASTIEN CHVALLIERES AMORIN CM.: "Validation of the insomnia severity index as an outcome measure for insomnia research", SLEEP MEDICINE, vol. 2, 2001, pages 297 - 307
BELANGER: "Rehabilitation After Traumatic Brain Injury", 2018, ELSEVIER, article "Community Reintegration", pages: 255
BELLON KMALEC JFKOLAKOWSKY-HAYNER SA.: "Mayo-portland adaptability inventory-4", THE JOURNAL OF HEAD TRAUMA REHABILITATION, vol. 27, 2012, pages 314 - 316
BERRY KJMIELKE JR PWIYER HK.: "Factorial designs and dummy coding", PERCEPTUAL AND MOTOR SKILLS, vol. 87, 1998, pages 919 - 927
BINESH NREZGHI M.: "Fuzzy clustering in community detection based on nonnegative matrix factorization with two novel evaluation criteria", APPLIED SOFT COMPUTING, vol. 69, 2018, pages 689 - 703
BOOKER ET AL.: "Predicting functional recovery after mild traumatic brain injury: the SHEFBIT cohort", BRAIN INJURY, vol. 33, no. 9, 2019, pages 1158 - 1165
BOSER ET AL.: "Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory", 1992, ACM PRESS, article "A training algorithm for optimal margin classifiers", pages: 142 - 152
BREIMAN: "Random Forests--Random Features", TECHNICAL REPORT 567, STATISTICS DEPARTMENT, September 1999 (1999-09-01)
BRUNET J-PTAMAYO PGOLUB TRMESIROV JP: "Metagenes and molecular pattern discovery using matrix factorization", PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, vol. 101, 2004, pages 4164 - 4169, XP009096387, DOI: 10.1073/pnas.0308531101
CAPLAIN ET AL.: "Efficacy of Psychoeducation and Cognitive Rehabilitation After Mild Traumatic Brain Injury for Preventing Post-concussional Syndrome in Individuals with High Risk of Poor Prognosis: A Randomized Clinical Trial", NEUROTRAM, vol. 10, 2019
CHIN ET AL.: "Improved Cognitive Performance Following Aerobic Exercise Training in People With Traumatic Brain Injury", ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION, vol. 96, no. 4, 2015, pages 754 - 759, XP029213187, DOI: 10.1016/j.apmr.2014.11.009
CNOSSEN MCSCHOLTEN ACLINGSMA HFSYNNOT AHAAGSMA JSTEYERBERG PEW ET AL.: "Predictors of major depression and posttraumatic stress disorder following traumatic brain injury: A systematic review and meta-analysis", THE JOURNAL OF NEUROPSYCHIATRY AND CLINICAL NEUROSCIENCES, vol. 29, 2017, pages 206 - 224
CORIS ET AL.: "Stimulant therapy utilization for neurocognitive deficits in mild traumatic brain injury", SPORTS HEALTH, vol. 14, no. 4, 2022, pages 538 - 548
CORRIGAN JDBOGNER J.: "Initial reliability and validity of the ohio state university TBI identification method", THE JOURNAL OF HEAD TRAUMA REHABILITATION, vol. 22, 2007, pages 318 - 329
DAHLBERG ET AL.: "Social communication skills training after traumatic brain injury", THE JOURNAL OF HEAD TRAUMA REHABILITATION, vol. 21, no. 5, 2006, pages 425
DAUSCHSAILIMAN: "Use of family focused therapy in rehabilitation for veterans with traumatic brain injury", REHABILITATION PSYCHOLOGY, vol. 54, no. 3, 2009, pages 279 - 287
DEROGATIS LRFITZPATRICK M., THE SCL-90-R, THE BRIEF SYMPTOM INVENTORY (BSI), AND THE BSI-18, 2004
DUDAHART: "Pattern Classification and Scene Analysis", 1973, JOHN WILEY & SONS, INC., pages: 211 - 256
ERTAS-SPANTGAR ET AL.: "Guiding patients with traumatic brain injury through the instrumental activities of daily living with the RehaGoal App: a feasibility study", DISABILITY AND REHABILITATION: ASSISTIVE TECHNOLOGY, vol. 19, no. 2, 2024, pages 254 - 65
ESSIG ET AL.: "Perfusion MRI: The Five Most Frequently Asked Technical Questions", AM. J. ROENTGENOL., vol. 200, 2013, pages 24 - 34, XP055455155, DOI: 10.2214/AJR.12.9543
FANN JRUOMOTO JMKATON WJ: "Cognitive improvement with treatment of depression following mild traumatic brain injury", PSYCHOSOMATICS, vol. 42, 2001, pages 48 - 54
FOKS KACNOSSEN MCDIPPEL DWMAAS AIMENON DNAALT J VAN DER ET AL.: "Management of mild traumatic brain injury at the emergency department and hospital admission in europe: A survey of 71 neurotrauma centers participating in the CENTER-TBI study", JOURNAL OF NEUROTRAUMA, vol. 34, 2017, pages 2529 - 2535
FUREY ET AL., BIOINFORMATICS, vol. 16, 2000, pages 906 - 914
GAUJOUX RSEOIGHE C.: "A flexible r package for nonnegative matrix factorization", BMC BIOINFORMATICS, vol. 11, 2010, pages 1 - 9
HAJEK CAYEATES KOGERRY TAYLOR HBANGERT BDIETRICH ANUSS KE ET AL.: "Relationships among post-concussive symptoms and symptoms of PTSD in children following mild traumatic brain injury", BRAIN INJURY, vol. 24, 2010, pages 100 - 109
HAMMOND FMCORRIGAN JDKETCHUM JMMALEC JFDAMS-O'CONNER KHART T ET AL.: "Prevalence of medical and psychiatric comorbidities following traumatic brain injury", THE JOURNAL OF HEAD TRAUMA REHABILITATION, vol. 34, 2019, pages E1
HARDINKELLY: "Seminars in Speech and Language", vol. 40, 2019, THIEME MEDICAL PUBLISHERS, article "The role of speech-language pathology in an interdisciplinary care model for persistent symptomatology of mild traumatic brain injury", pages: 065 - 078
HART TFANN JRCHERVONEVA IJUENGST SBROSENTHAL JAKRELLMAN JW ET AL.: "Prevalence, risk factors, and correlates of anxiety at 1 year after moderate to severe traumatic brain injury", ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION, vol. 97, 2016, pages 701 - 707, XP029522463, DOI: 10.1016/j.apmr.2015.08.436
HASSOUN: "Fundamentals of Artificial Neural Networks", MASSACHUSETTS INSTITUTE OF TECHNOLOGY, 1995
HASTIE TQIAN JTAY K.: "An introduction to glmnet", CRAN R REPOSITARY, 2021
HELLEWELL SCBEATON CSWELTON TGRIEVE SM.: "Characterizing the risk of depression following mild traumatic brain injury: A meta-analysis of the literature comparing chronic mTBI to non-mTBI populations", FRONTIERS IN NEUROLOGY, vol. 11, 2020, pages 350
HOERL AEKENNARD RW: "Ridge regression: Biased estimation for nonorthogonal problems", TECHNOMETRICS, vol. 12, 1970, pages 55 - 67
HOFFERXBALABANY: "Vestibular rehabilitation after mild traumatic brain injury with vestibular pathology", NEUROREHABILITATION, vol. 29, no. 2, 2011, pages 167 - 171
HOGE CWCASTRO CA: "Treatment of generalized war-related health concerns: Placing TBI and PTSD in context", JAMA, vol. 312, 2014, pages 1685 - 1686
HONG FTIAN LDEVANARAYAN V.: "Improving the robustness of variable selection and predictive performance of regularized generalized linear models and cox proportional hazard models", MATHEMATICS, vol. 11, 2023, pages 557
KAKADEKANASE: "Effect of Multidimensional Exercise Program for Improving Balance in Traumatic Brain Injury Patients", MEDICO-LEGAL UPDATE, vol. 20, no. 3, 2020, pages 155 - 161
KRIZHEVSKY ET AL.: "Advances in Neural Information Processing Systems", vol. 2, 2012, CURRAN ASSOCIATES, INC., article "Imagenet classification with deep convolutional neural networks", pages: 1097 - 1105
KROENKE ET AL.: "The Patient Health Questionnaire-2: validity of a two-item depression screener", MED CARE, vol. 41, no. 11, 2003, pages 1284 - 92
KROENKE KSPITZER RLWILLIAMS JB: "The PHQ-9: Validity of a brief depression severity measure", JOURNAL OF GENERAL INTERNAL MEDICINE, vol. 16, 2001, pages 606 - 613
KULBE JRJAIN SNELSON LDKORLEY FKMUKHERJEE PSUN X ET AL.: "Association of day-of-injury plasma glial fibrillary acidic protein concentration and six-month posttraumatic stress disorder in patients with mild traumatic brain injury", NEUROPSYCHOPHARMACOLOGY, vol. 47, 2022, pages 2300 - 2308
LANGLOIS JARUTLAND-BROWN WWALD MM: "The epidemiology and impact of traumatic brain injury: A brief overview", THE JOURNAL OF HEAD TRAUMA REHABILITATION, vol. 21, 2006, pages 375 - 378
LAROCHELLE ET AL.: "Exploring strategies for training deep neural networks", J MACH LEARN RES, vol. 10, 2009, pages 1 - 40
LAZAR CDONCESCU A.: "2009 international conference on complex, intelligent and software intensive systems", 2009, IEEE, article "Non negative matrix factorization clustering capabilities; application on multivariate image segmentation", pages: 924 - 929
LE BIHAN ET AL.: "Diffusion tensor imaging: Concepts and applications", J. MAGN. RESON. IMAGING, vol. 13, 2001, pages 534 - 546, XP002358554, DOI: 10.1002/jmri.1076
LEE DDSEUNG HS.: "Learning the parts of objects by non-negative matrix factorization", NATURE, vol. 401, 1999, pages 788 - 791, XP008056832, DOI: 10.1038/44565
LEVIN HSDIAZ-ARRASTIA RR.: "Diagnosis, prognosis, and clinical management of mild traumatic brain injury", THE LANCET NEUROLOGY, vol. 14, 2015, pages 506 - 517
LOGOTHETIS ET AL.: "Neurophysiological investigation of the basis of the fMRI signal", NATURE, vol. 412, 2001, pages 259,262 - 408,411-412
LOVETTE ET AL.: "Hidden gains''? Measuring the impact of mindfulness-based interventions for people with mild traumatic brain injury: a scoping review", BRAIN INJURY, vol. 36, no. 9, 2022, pages 1059 - 1070
MAAS ANDREW I ET AL: "Traumatic brain injury: progress and challenges in prevention, clinical care, and research", THE LANCET NEUROLOGY, vol. 21, no. 11, 29 September 2022 (2022-09-29), AMSTERDAM, NL, pages 1004 - 1060, XP093177791, ISSN: 1474-4422, DOI: 10.1016/S1474-4422(22)00309-X *
MAC DONALD CLBARBER JJORDAN MJOHNSON AMDIKMEN SFANN JR ET AL.: "Early clinical predictors of 5-year outcome after concussive blast traumatic brain injury", JAMA NEUROLOGY, vol. 74, 2017, pages 821 - 829
MACLENNAN: "Cognitive-communication rehabilitation for combat-related mild traumatic brain injury", JOURNAL OF REHABILITATION RESEARCH AND DEVELOPMENT, vol. 49, no. 7, 2012, pages XI
MAHADEWA ET AL.: "Modified Revised Trauma-Marshall score as a proposed tool in predicting the outcome of moderate and severe traumatic brain injury", EMERG MED., vol. 10, 2018, pages 135 - 139
MALLYA SSUTHERLAND JPONGRACIC SMAINLAND BORNSTEIN TJ.: "The manifestation of anxiety disorders after traumatic brain injury: A review", JOURNAL OF NEUROTRAUMA, vol. 32, 2015, pages 411 - 421
MARCOLINIHINE: "Approach to the Diagnosis and Management of Subarachnoid Hemorrhage", WEST J EMERG MED, vol. 20, no. 2, 2019, pages 203 - 211
MCCREA MGUSKIEWICZ KMMARSHALL SWBARR WRANDOLPH CCANTU RC ET AL.: "Acute effects and recovery time following concussion in collegiate football players: The NCAA concussion study", JAMA, vol. 290, 2003, pages 2556 - 2563
MCLACHLAN ET AL., BIOINFORMATICS, vol. 18, no. 3, 2002, pages 413 - 422
MCMAHON ET AL.: "Symptomatology and Functional Outcome in Mild Traumatic Brain Injury: Results from the Prospective TRACK-TBI Study", JOURNAL OF NEUROTRAUMA, vol. 31, 2014, pages 26 - 33
MCMILLAN TWILSON LPONSFORD JLEVIN HTEASDALE GBOND M.: "The glasgow outcome scale-40 years of application and refinement", NATURE REVIEWS NEUROLOGY, vol. 12, 2016, pages 477 - 485
MICHOPOULOS VROTHBAUM AOJOVANOVIC TALMLI LMBRADLEY BROTHBAUM BO ET AL.: "Association of CRP genetic variation and CRP level with elevated PTSD symptoms and physiological responses in a civilian population with high levels of trauma", AMERICAN JOURNAL OF PSYCHIATRY, vol. 172, 2015, pages 353 - 362
MIKOLIC ET AL.: "Prediction of Global Functional Outcome and Post-Concussive Symptoms after Mild Traumatic Brain Injury: External Validation of Prognostic Models in the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study", JOURNAL OF NEUROTRAUMA, 2020, pages 196 - 209
MIKOLIC ET AL.: "Prognostic Models for Global Functional Outcome and Post-Concussion Symptoms Following Mild Traumatic Brain Injury: A Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study", JOURNAL OF NEUROTRAMA, August 2023 (2023-08-01), pages 1651 - 1670
MILLERMITTENBERG: "Brief Cognitive Behavioral Interventions in Mild Traumatic Brain Injury", APPLIED NEUROPSYCHOLOGY, vol. 5, no. 4, 2010, pages 172 - 183
MOHAMMADIFARD ET AL.: "Marshall and Rotterdam Computed Tomography scores in predicting early deaths after brain trauma", EUR J TRANSL MYOL, vol. 28, no. 3, 2008, pages 265 - 273
MOONJEON: "Effects of Virtual Reality-Based Activities of Daily Living Training on Activities of Daily Living and Rehabilitative Motivation in Patients with Traumatic Brain Injury: A Pilot Study", THERAPEUTIC SCIENCE FOR REHABILITATION, vol. 8, no. 4, 2019, pages 41 - 51
MOORE ELTERRYBERRY-SPOHR LHOPE DA: "Mild traumatic brain injury and anxiety sequelae: A review of the literature", BRAIN INJURY, vol. 20, 2006, pages 117 - 132
NADEL: "Emerging Utility of Applied Magnetic Resonance Imaging in the Management of Traumatic Brain Injury", MED. SCI., vol. 9, no. 10, 2021
NAWABI ET AL.: "Non-contrast computed tomography features predict intraventricular hemorrhage growth", EUROPEAN RADIOLOGY, vol. 33, pages 7807 - 7817
NELSON LDFURGER RERANSON JTARIMA SHAMMEKE TARANDOLPH C ET AL.: "Acute clinical predictors of symptom recovery in emergency department patients with uncomplicated mild traumatic brain injury or non-traumatic brain injuries", JOURNAL OF NEUROTRAUMA, vol. 35, 2018, pages 249 - 259
NELSON LDTEMKIN NRDIKMEN SBARBER JGIACINO JTYUH E ET AL.: "Recovery after mild traumatic brain injury in patients presenting to US level i trauma centers: A transforming research and clinical knowledge in traumatic brain injury (TRACK-TBI) study", JAMA NEUROLOGY, vol. 76, 2019, pages 1049 - 1059
NG ET AL.: "On discriminative vs. generative classifiers: A comparison of logistic regression and naive Bayes", ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, vol. 14, 2002
PATTINSON CLSHAHIM PTAYLOR PDUNBAR KGUEDES VMOTAMEDI V ET AL.: "Elevated tau in military personnel relates to chronic symptoms following traumatic brain injury", THE JOURNAL OF HEAD TRAUMA REHABILITATION, vol. 35, 2020, pages 66
PAVOT WDIENER E.: "The satisfaction with life scale and the emerging construct of life satisfaction", THE JOURNAL OF POSITIVE PSYCHOLOGY, vol. 3, 2008, pages 137 - 152
PETCHARUNPAISAN ET AL.: "Arterial spin labeling in neuroimaging", WORLD J. RADIOL., vol. 2, 2010, pages 384 - 398
PETRELLAPROVENZALE: "MR Perfusion Imaging of the Brain: Techniques and applications", AM. J. ROENTGENOL., vol. 175, 2000, pages 207 - 219, XP008013101
PONSFORD JCAMERON PFITZGERALD MGRANT MMIKOCKA-WALUS ASCHONBERGER M.: "Predictors of postconcussive symptoms 3 months after mild traumatic brain injury", NEUROPSYCHOLOGY, vol. 26, 2012, pages 304
RADHAKRISHNAN RGARAKANI AGROSS LSGOIN MKPINE JSLABY AE ET AL.: "Neuropsychiatric aspects of concussion", THE LANCET PSYCHIATRY, vol. 3, 2016, pages 1166 - 1175
RAPOPORT MJMCCULLAGH SSTREINER DFEINSTEIN A.: "The clinical significance of major depression following mild traumatic brain injury", PSYCHOSOMATICS, vol. 44, 2003, pages 31 - 37
ROGERS JMREAD CA: "Psychiatric comorbidity following traumatic brain injury", BRAIN INJURY, vol. 21, 2007, pages 1321 - 1333
ROHLING MLBINDER LMDEMAKIS GJLARRABEE GJPLOETZ DMLANGHINRICHSEN-ROHLING J. ET AL.: "A meta-analysis of neuropsychological outcome after mild traumatic brain injury: Re-analyses and reconsiderations of binder et al., frencham et al., and pertab et al.", THE CLINICAL NEUROPSYCHOLOGIST., vol. 25, 2011, pages 608 - 623
RUMELHART ET AL.: "ch. Learning Representations by Back-propagating Errors", 1988, MIT PRESS, article "Neurocomputing: Foundations of research", pages: 696 - 699
SCHLIEP ET AL., BIOINFORMATICS, vol. 19, no. 1, 2003, pages i255 - i263
SHAHAPURE KRNICHOLAS C.: "2020 IEEE 7th international conference on data science and advanced analytics (DSAA", 2020, IEEE, article "Cluster quality analysis using silhouette score", pages: 747 - 748
SILVERBERG NOAH D. ET AL: "Systematic Review of Multivariable Prognostic Models for Mild Traumatic Brain Injury", JOURNAL OF NEUROTRAUMA., vol. 32, no. 8, 15 April 2015 (2015-04-15), US, pages 517 - 526, XP093177798, ISSN: 0897-7151, DOI: 10.1089/neu.2014.3600 *
SMITH ET AL.: "Advanced neuroimaging in traumatic brain injury: An overview", NEUROSURG. FOCUS, vol. 47, 2019, pages E17
TANEV KSPENTEL KZKREDLOW MACHARNEY ME: "PTSD and TBI co-morbidity: Scope, clinical presentation and treatment options", BRAIN INJURY, vol. 28, 2014, pages 261 - 270
TAYLOR CABELL JMBREIDING MJXU L.: "Traumatic brain injury-related emergency department visits, hospitalizations, and deaths-united states, 2007 and 2013", MMWR SURVEILLANCE SUMMARIES, vol. 66, no. 1, 2017
TEASDALE GMAAS ALECKY FMANLEY GSTOCCHETTI NMURRAY G.: "The glasgow coma scale at 40 years: Standing the test of time", THE LANCET NEUROLOGY, vol. 13, 2014, pages 844 - 854
TEASDALEJENNETT: "Assessment of coma and impaired consciousness: A practical scale", LANCET, vol. 304, 1974, pages 81 - 84
TELISCHAK ET AL.: "Arterial spin labeling MRI: Clinical applications in the brain", J. MAGN. RESON. IMAGING, vol. 41, 2015, pages 1165 - 1180
THIAGARAJANCIUFFREDA: "Effect of oculomotor rehabilitation on vergence responsivity in mild traumatic brain injury", JRRD, vol. 50, no. 9, 2013, pages 1223 - 1240
THOMPSON HJVAVILALA MSRIVARA FP: "Common data elements and federal interagency traumatic brain injury research informatics system for TBI research", ANNU REV NURS RES., vol. 33, 2015, pages 1 - 11
TIBSHIRANI R.: "Regression shrinkage and selection via the lasso", JOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES B (METHODOLOGICAL, vol. 58, 1996, pages 267 - 288
VAN DER MAATEN LHINTON G.: "Visualizing data using t-SNE", JOURNAL OF MACHINE LEARNING RESEARCH, vol. 9, 2008
VAN DER NAALT JOUKJE ET AL: "Early predictors of outcome after mild traumatic brain injury (UPFRONT): an observational cohort study", THE LANCET NEUROLOGY, ELSEVIER, AMSTERDAM, NL, vol. 16, no. 7, 13 June 2017 (2017-06-13), pages 532 - 540, XP085075188, ISSN: 1474-4422, DOI: 10.1016/S1474-4422(17)30117-5 *
VIDHYA ET AL.: "Automated Detection and Screening of Traumatic Brain Injury (TBI) Using Computed Tomography Images: A Comprehensive Review and Future Perspectives", INT. J. ENVIRON. RES. PUBLIC HEALTH, vol. 18, 2021, pages 6499
VINCENT ET AL.: "Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion", J MACH LEARN RES, vol. 11, 2010, pages 3371 - 3408
VINK GFRANK LEPANNEKOEK JVAN BUUREN S.: "Predictive mean matching imputation of semicontinuous variables", STATISTICA NEERLANDICA, vol. 68, 2014, pages 61 - 90
WANG Y-XZHANG Y-J.: "Nonnegative matrix factorization: A comprehensive review", IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, vol. 25, 2012, pages 1336 - 1353
WILDE ELISABETH A ET AL: "A Framework to Advance Biomarker Development in the Diagnosis, Outcome Prediction, and Treatment of Traumatic Brain Injury", JOURNAL OF NEUROTRAUMA., vol. 39, no. 7-8, 11 May 2017 (2017-05-11), US, pages 436 - 457, XP093177777, ISSN: 0897-7151, Retrieved from the Internet <URL:https://www.liebertpub.com/doi/full/10.1089/neu.2021.0099> DOI: 10.1089/neu.2021.0099 *
XU ET AL.: "Complementary and alternative interventions for fatigue management after traumatic brain injury: a systematic review", THERAPEUTIC ADVANCES IN NEUROLOGICAL DISORDERS, vol. 10, no. 5, 2017, pages 229 - 239
XU ET AL.: "High-Sensitivity C-Reactive Protein Is a Prognostic Biomarker of Six-Month Disability after Traumatic Brain Injury: Results from the TRACK-TBI Study", JOURNAL OF NEUROTRAUMA, vol. 38, 2021, pages 918 - 927
XU LBYUE JKKORLEY FPUCCIO AMYUH ELSUN X ET AL.: "High-sensitivity c-reactive protein is a prognostic biomarker of six-month disability after traumatic brain injury: Results from the TRACK-TBI study", JOURNAL OF NEUROTRAUMA, vol. 38, 2021, pages 918 - 927
YUE ET AL.: "Selective serotonin reuptake inhibitors for treating neurocognitive and neuropsychiatric disorders following traumatic brain injury: an evaluation of current evidence", BRAIN SCIENCES, vol. 7, no. 8, 2017, pages 93
YUH ELJAIN SSUN XPISICA DHARRIS MHTAYLOR SR ET AL.: "Pathological computed tomography features associated with adverse outcomes after mild traumatic brain injury: A TRACK-TBI study with external validation in CENTER-TBI", JAMA NEUROLOGY, vol. 78, 2021, pages 1137 - 1148
ZEILER: "ADADELTA: an adaptive learning rate method", CORR, vol. abs/1212.5701, 2012
ZHANG SWANG R-SZHANG X-S: "Uncovering fuzzy community structure in complex networks", PHYSICAL REVIEW E, vol. 76, 2007, pages 046103
ZHANG Z.: "Multiple imputation with multivariate imputation by chained equation (MICE) package", ANNALS OF TRANSLATIONAL MEDICINE, vol. 4, 2016
ZOU HHASTIE T.: "Regularization and variable selection via the elastic net", JOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES B (STATISTICAL METHODOLOGY, vol. 67, 2005, pages 301 - 320, XP055849418, DOI: 10.1111/j.1467-9868.2005.00503.x

Similar Documents

Publication Publication Date Title
Zhu et al. Association of prepubertal and postpubertal exposure to childhood maltreatment with adult amygdala function
Baniqued et al. Brain network modularity predicts exercise-related executive function gains in older adults
Weinberg et al. Cognitive subtypes of schizophrenia characterized by differential brain volumetric reductions and cognitive decline
Asarnow et al. Neurocognitive impairments in nonpsychotic parents of children with schizophrenia and attention-deficit/hyperactivity disorder: the University of California, Los Angeles Family Study
Kessler et al. Growth charting of brain connectivity networks and the identification of attention impairment in youth
Dean et al. Motor clusters reveal differences in risk for psychosis, cognitive functioning, and thalamocortical connectivity: evidence for vulnerability subtypes
Dickinson et al. Overlooking the obvious: a meta-analytic comparison of digit symbol coding tasks and other cognitive measures in schizophrenia
Ehlers et al. Regional brain volumes moderate, but do not mediate, the effects of group-based exercise training on reductions in loneliness in older adults
Miller et al. Neuroanatomical correlates of religiosity and spirituality: a study in adults at high and low familial risk for depression
Tamboer et al. Machine learning and dyslexia: Classification of individual structural neuro-imaging scans of students with and without dyslexia
Haider et al. Association between hospitals caring for a disproportionately high percentage of minority trauma patients and increased mortality: a nationwide analysis of 434 hospitals
Tanaka et al. The brain basis of the phonological deficit in dyslexia is independent of IQ
Mun et al. Finding groups using model-based cluster analysis: heterogeneous emotional self-regulatory processes and heavy alcohol use risk.
Hoeft et al. Morphometric spatial patterns differentiating boys with fragile X syndrome, typically developing boys, and developmentally delayed boys aged 1 to 3 years
Langmaid et al. Handwriting in children with ADHD
US20190167179A1 (en) Methods and system for assessing a cognitive function
Baum et al. A comparison of WISC-IV and SB-5 intelligence scores in adolescents with autism spectrum disorder
Wojtowicz et al. Altered functional connectivity and performance variability in relapsing–remitting multiple sclerosis
Zhang et al. Clinical subtypes that predict conversion to psychosis: a canonical correlation analysis study from the ShangHai At Risk for Psychosis program
Li et al. A face image classification method of autistic children based on the two-phase transfer learning
Erhan et al. Obsessive compulsive features predict cautious decision strategies
Schubart et al. Use of cluster analysis to delineate symptom profiles in an Ehlers-Danlos syndrome patient population
Park et al. Clinical use of continuous performance tests to diagnose children with ADHD
Gullett et al. Baseline neuroimaging predicts decline to dementia from amnestic mild cognitive impairment
Leister et al. Routine blood chemistry predicts functional recovery after traumatic spinal cord injury: a post hoc analysis

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24719329

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE