[go: up one dir, main page]

WO2017087680A1 - Système de détection de crise et d'atténuation des blessures et son procédé d'utilisation - Google Patents

Système de détection de crise et d'atténuation des blessures et son procédé d'utilisation Download PDF

Info

Publication number
WO2017087680A1
WO2017087680A1 PCT/US2016/062543 US2016062543W WO2017087680A1 WO 2017087680 A1 WO2017087680 A1 WO 2017087680A1 US 2016062543 W US2016062543 W US 2016062543W WO 2017087680 A1 WO2017087680 A1 WO 2017087680A1
Authority
WO
WIPO (PCT)
Prior art keywords
sensor
data
signals
vestibular
user
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.)
Ceased
Application number
PCT/US2016/062543
Other languages
English (en)
Inventor
Allon Guez
Michael R. SPERLING
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.)
Gracefall Inc
Original Assignee
Gracefall 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
Priority claimed from US15/269,820 external-priority patent/US9974344B2/en
Application filed by Gracefall Inc filed Critical Gracefall Inc
Publication of WO2017087680A1 publication Critical patent/WO2017087680A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • 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
    • 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

Definitions

  • the present invention relates to a system for detection of seizures, such as epileptic seizures. It also relates to a system for protecting a person from injuries that may otherwise be sustained by a fall, collision, or other movement with the potential for causing injury.
  • Epilepsy is a chronic neurological disorder, which is characterized by recurrent seizures.
  • An epileptic seizure is the result of abnormal electrical activity in the brain caused by hypersynchronous and abnormal discharges of neurons in the brain.
  • Convulsions known as tonic-clonic seizures, produce loss of consciousness, whole body stiffening and then repetitive contractions of muscles (NIND, N.I.o.N.D.a.S.-. Epilepsy: Hope Through Research. National Institution of Health, 2015).
  • NIND N.I.o.N.D.a.S.-. Epilepsy: Hope Through Research. National Institution of Health, 2015.
  • According to the World Health Organization Report in 2015 approximately 50 million people worldwide have epilepsy, making it one of the most common neurological diseases.
  • EEG, MRS, and other neural-imaging techniques are becoming more useful in being able to detect certain states of mind prior to the subject actually knowing how he or she feels and whether a fall or a disorder, such as an epileptic seizure, is imminent. It is known that brain signals can be used to indicate a person's current activity state and to predict a change in the person's activity state. These techniques can be used for event detection (i.e., the early detection of falls, startling reflexes).
  • seizure detection devices which are on, or planned to be on the market, as representatively shown in Table 1.
  • the first 6 devices are intended to be used at home.
  • the limitations of these devices include high false positive rates and low sensitivity.
  • the 7th device is used in hospital settings for EEG recording and seizure detection. For this device, the false positive rate can be high as 10 times/day. There is an evident need for improvement in seizure detection to improve accuracy and reduce false- positive rates.
  • Emfit Movement EMFIT Bed Sensor Vehicle Sensor
  • prior art systems face problems associated with the timing and sensitivity of detecting falls and other movement-related events which can lead to injury.
  • the prior art systems may not be able to detect a fall/startling movement with confidence until there is not enough time to deploy a safety device to prevent injury during the fall/startling movement.
  • This may be in part due to the fact that the prior art systems typically depend entirely on accelerometers or gyroscopes to detect and respond to movement. Signals obtained from the accelerometers or gyroscopes which indicate a fall/startling movement may be available only after the fall is already initiated and in progress. Accordingly, there may not be enough time after a fall is detected from signals obtained from the accelerometers or gyroscopes to deploy the safety device.
  • Prior art systems may also suffer from a high degree of sensitivity leading to a high number of false detections of potential injury events (i.e., false positives). This is due in part to prior art systems relying on information solely from gyroscopes and accelerometers. In signals acquired from gyroscopes and accelerometers intentional movements such as laughing, dancing, jumping, turns, and the like may have similar motion profiles as the motion profiles associated with unintended movement such as falls and startling effects. Accordingly, in prior art systems which solely utilized gyroscopes and accelerometers it would be impossible to distinguish between intentional and unintentional movements, leading to a high number of false positives.
  • the present invention provides a system and a method for detecting a seizure, especially an epileptic seizure.
  • the invention also provides a system for detecting body signals indicative of an movement event such as an imminent fall and/or a startling movement and activating a safety device to mitigate damage or injury resulting from the fall and/or startling movement.
  • the system and method are capable of adapting its performance to the individual user.
  • the body signals can be indicative of activity in the vestibular (and startling/surprise related) system discovered to provide timely and accurate information regarding an imminent fall and/or startling movement.
  • the vestibular (and startling/ surprise related) system is particularly sensitive to movement, it has long been the focus of stimulation techniques rather than motion detection as is used herein.
  • An exemplary embodiment of the present invention includes methods and systems having at least one sensor configured to obtain one or more signals from a user, wherein the at least one sensor comprises a sensor for detecting epileptic seizures.
  • the system may further comprise a vestibular (and startling/surprise related) sensor specially configured to obtain vestibular (and startling/surprise related) signals.
  • the exemplary embodiment of the present invention also includes at least one processor electronically coupled to the at least one sensor and to a memory storing computer readable instructions which cause the processor receive the one or more signals from the at least one sensor, extract vestibular (and startling/surprise related) data indicative of a state of the user's vestibular (and startling/surprise related) system from the one or more signals, determine whether the user is undergoing an event by comparing the vestibular (and startling/surprise related) data to predetermined data, and transmit an activation signal to a safety device when it is determined that the user is undergoing the event.
  • the methods and systems further include a safety device configured to deploy when the activation signal is received from the at least one processor.
  • FIG. 1 shows a schematic drawing of a seizure detection system according to an exemplary embodiment of the present invention.
  • FIG. 2 shows a schematic drawing of a fall and collision detection and injury mitigation system according to an exemplary embodiment of the present invention.
  • FIG. 3 shows a deployable airbag that can be used with the fall and collision detection and injury mitigation system of FIGS. 2 and 6.
  • FIG. 4 shows an exoskeleton that can be used with the fall and collision detection and injury mitigation system of FIGS. 2 and 6.
  • FIG. 5 is a flowchart describing an exemplary signal and information flow as well as exemplary modules for providing adaptation and tuning capabilities of the system for individual users.
  • FIG. 6 shows a schematic drawing of a fall and collision detection and injury mitigation system using a vestibular (and startling/surprise related) motion sensor according to an exemplary embodiment of the present invention.
  • FIG. 7A shows a vestibular (and startling/surprise related) motion sensor in connection with a hearing aid-like device located behind a user's ear according to an exemplary embodiment of the present invention.
  • FIG. 7B shows a vestibular (and startling/surprise related) motion sensor in connection with a hearing aid-like device inserted in a user's ear according to another exemplary embodiment of the present invention.
  • FIG. 8 shows a vestibular (and startling/surprise related) motion sensor in connection with a set of glasses according to an exemplary embodiment of the present invention.
  • FIG. 9 shows an exemplary wearable safety device for use with the vestibular (and startling/surprise related) motion sensors according to an exemplary embodiment of the present invention.
  • FIG. 10 shows example data obtained from vestibular (and startling/surprise related) sensors used in connection with an exemplary embodiment of the present invention.
  • a component may be, but is not limited to being, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and/or a computer.
  • an application running on a controller and the controller can be a component.
  • One or more components may reside within a process and/or thread of execution and a component may be localized on one computer and/or distributed between two or more computers.
  • the present invention may be implemented as circuit-based processes, including possible implementation as a single integrated circuit (such as an ASIC or an FPGA), a multi- chip module, a single card, or a multi-card circuit pack.
  • various functions of circuit elements may also be implemented as processing blocks in a software program.
  • Such software may be employed in, for example, a digital signal processor, micro-controller, or general-purpose computer.
  • the present invention can be embodied in the form of methods and apparatuses for practicing those methods.
  • the present invention can also be embodied in the form of program code embodied in tangible media, such as magnetic recording media, optical recording media, solid state memory, floppy diskettes, CD-ROMs, hard drives, or any other machine- readable storage medium, wherein, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention.
  • the present invention can also be embodied in the form of program code, for example, whether stored in a storage medium, loaded into and/or executed by a machine, or transmitted over some transmission medium or carrier, such as over electrical wiring or cabling, through fiber optics, or via electromagnetic radiation, wherein, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention.
  • the program code segments When implemented on a general -purpose processor, the program code segments combine with the processor to provide a unique device that operates analogously to specific logic circuits.
  • the present invention can also be embodied in the form of a bitstream or other sequence of signal values electrically or optically transmitted through a medium, stored magnetic-field variations in a magnetic recording medium, etc., generated using a method and/or an apparatus of the present invention.
  • the term "compatible" means that the element communicates with other elements in a manner wholly or partially specified by the standard, and would be recognized by other elements as sufficiently capable of communicating with the other elements in the manner specified by the standard. The compatible element does not need to operate internally in a manner specified by the standard.
  • Couple refers to any manner known in the art or later developed in which energy is allowed to be transferred between two or more elements, and the interposition of one or more additional elements is contemplated, although not required. Conversely, the terms “directly coupled,” “directly connected,” etc., imply the absence of such additional elements.
  • a seizure detection system is graphically represented by Figure 1.
  • the invention provides an integrated epileptic seizure
  • the system consists of an array of sensors including (but not limited to) EEG, EMG, EKG, accelerometer, gyroscope, video camera, audio, eye movement, galvanic skin response, and temperature which read a variety of body physiological and kinematic and dynamic signals.
  • the system exploits the synergism within the array of sensors to provide enhanced seizure detection followed by enhanced movement detection.
  • the system also includes a central processing unit which assembles and processes the data from all the sensors and executes a customized algorithm with machine learning in several modules.
  • the seizure dectection module will be used to alert the person and possibly others (for healthcare and safety etc.) that a seizure is or will be occurring. All references noted below are incorporated into this disclosure by this reference for purposes of providing context to the invention.
  • Wavelet transform is one of the techniques used to analyze the EEG signal. Wavelet transform is particularly effective for representing various aspects of non- stationary signals such as trends, discontinuities, and repeated patterns, where other signal processing approaches fail or are not as effective (Adeli, H., Z. Zhou, and N. Dadiolo, Analysis of EEG records in an epileptic patient using wavelet transform. Journal of Neuroscience Methods, 2003. 123(1): p. 69-87). Entropy is also a good candidate in analysis of EEG which is highly non-linear and non-stationary. In prior research (U. Rajendra Acharya, F.M., S.
  • ECG abnormalities occur often and repeatedly in several seizures of the same patient (Maeike Zijlmans, D.F., Jean Gotman, Heart Rate Changes and ECG Abnormalities During Epileptic Seizure: Prevalence and Definition of an Objective Clinical Sign. 2002).
  • Electrodermal activity is a sensitive index of sympathetic nervous system activity (Poh, M.-Z., N.C. Swenson, and R.W. Picard, A Wearable Sensor for Unobtrusive, Long-Term Assessment of Electrodermal Activity. IEEE Transactions on Biomedical Engineering, 2010. 57(5): p. 1243-1252). Epileptic seizures induce a surge in EDA. These changes are greater in generalized tonic- clonic seizures and reflect a massive sympathetic discharge.
  • Kalman filtering also known as linear quadratic estimation (LQE)
  • LQE linear quadratic estimation
  • the Kalman filter does not require any assumption that the errors are Gaussian (Kalman, R.E., A New
  • Sensor fusion techniques provide improved and robust estimates. It is known that fusion of multiple signals that are correlated with the same process parameter can estimate that parameter better. This is because different signals have different correlation efficiency and their effective and cooperative fusion is expected to produce a better estimation result.
  • Machine learning approaches seek to incorporate multiple data points in order to build a model that can make predictions in future data sets (Rudie, J.D., J.B. Colby, and N. Salamon, Machine learning classification of mesial temporal sclerosis in epilepsy patients. Epilepsy Research, 2015. 117: p. 63-69). Machine learning explores the study and construction of algorithms that can learn from and make predictions on data. Such algorithms operate by building a model from example inputs in order to make data-driven predictions or decisions, rather than following strictly static program instructions.
  • EEG seizure detection uses various characteristics of the EEG signal including changes in frequency, amplitude, rhythmicity and spike rate from the precital state to determine whether a seizure has begun.
  • existing detection algorithms lack sufficient specificity for reliable detection for purposes of activating an automatic protection system.
  • Use of other physiological signals greatly improves sensitivity and specificity.
  • EKG EKG
  • EMG electrodermal activity (galvanic skin response), eye movement, and patterns of limb and body movement, among others.
  • Seizures were captured during continuous video-electroencephalogram (EEG) monitoring that allowed precise correlation between clinical manifestations and EEG changes[18] Yu, H.-Y., et al., Lateralizing value of early head turning and ictal dystonia in temporal lobe seizures: a video-EEG study. Seizure, 2001. 10(6): p. 428-432. Early head turning occurred in 222 seizures (84.7%) from 77 patients[18] Yu, H.-Y., et al., Lateralizing value of early head turning and ictal dystonia in temporal lobe seizures: a video-EEG study. Seizure, 2001. 10(6): p. 428-432.
  • Root Square Mean function is used to characterize the amplitude of electrical signals from the musculature: [0056] N-l
  • MF is defined as the frequency that divides the magnitude spectrum in two parts of equal sizes (the area under the curve for the frequencies lower than MF equals the area under the curve for the frequencies higher than MF), and it is expressed according to the formula:
  • IIFFTm(f)I 0.5
  • IIFFTm(f)I,m 1,2,3
  • n is the window number
  • fs is the sampling frequency
  • fMF is the MF
  • FFTm is the discrete frequency spectrum of the window m.
  • Automatisms are non-purposeful, stereotyped, and repetitive behaviors that commonly accompany complex partial seizures.
  • the most common automatisms, at least in temporal lobe epilepsy, are oral (e.g., lip smacking, chewing, swallowing) and manual (eg, picking, fumbling, patting). Falls are much less common in these seizures.
  • EEG signals is combined with additional sensor data for fall and collision detection in a seizure-vector field to implement a multiple machine learning algorithm to detect or even predict seizure.
  • System 100 uses physiological and movement data to anticipate when a person (“the user") wearing system 100 is in imminent danger or is in fact falling and activates an injury mitigation device to attempt to minimize potential injury from the fall or to prevent fall altogether.
  • System 100 is comprised of a plurality of sensors attached to the user that measure physiological and physical data and transmit that data to a processor, which in turn determines whether the data are indicative of an imminent fall. If so, the processor transmits an electronic signal to activate the injury mitigation device.
  • System 100 includes a brain sensor 110 that is adapted to receive a first electrical signal and to transmit a first electronic signal based on the first electrical signal.
  • brain sensor 110 can be an electro-encephalocardiogram ("EEG") sensor, a near infrared sensor (NIRS), or other known or as yet unknown type of brain sensor.
  • EEG electro-encephalocardiogram
  • NIRS near infrared sensor
  • brain sensor 110 can be attached (invasively or non-invasively based on the brain sensor type) to the user's forehead, skull, top of the neck or back of the head depending on the specific neural tissue such as, for example, the vestibular (and startling/surprise related) system, Central Nervous System or peripheral nervous system that is being tracked.
  • System 100 also includes a muscular sensor 120 that is adapted to receive a second electrical signal and to transmit a second electronic signal based on the second electrical signal.
  • muscular sensor can be an electromyography ("EMG") sensor, or other known or as yet unknown type of muscular sensor.
  • EMG electromyography
  • muscular sensor 120 can be attached to the user's thighs, picking up sensory information from quadriceps or hamstrings; upper arms; neck; or other muscle groups whose reflexive electrical activation is being pursued and monitored as related to vestibular (and startling/surprise) response.
  • System 100 also includes a movement sensor 130 that is adapted to sense movement and to transmit a third electronic signal based on the movement.
  • movement sensor 130 can be an accelerometer, gyro or other known or as yet unknown type of velocity, displacement, acceleration, jerk or other movement sensor.
  • movement sensor 130 can be attached to the user's chest. Sensor 130, however, may be located anywhere on the user's core and head, depending on the product design, convenience, and other potential user disabilities.
  • FIG. 2 shows brain sensor 110, muscular sensor 120, and movement sensor 130 as three separate sensors
  • brain sensor 110, muscular sensor 120, and movement sensor 130 can be provided as a single unit 112 that needs only to be attached to a single location on the user.
  • An exemplary location for single unit 112 can be on the back of the user's neck.
  • brain sensor 110 is able to sense neurological signals generated by the vestibular nuclei, on either side of the brain stem.
  • Muscular sensor 120 can alternatively be used to sense movement of the trapezius and sternocleidomastoid muscles of the neck and movement sensor 130 can sense movement of the user's body.
  • a processor 140 is electronically coupled to brain signal sensor 110, muscular signal sensor 120, and movement sensor 130.
  • Processor 140 is configured to process the first electronic signal, the second electronic signal, and the third electronic signal and generate a result.
  • Processor 140 can be an electronic microprocessor and is powered by an electrical power source, such as a battery (not shown).
  • Processor 140 can include a memory 142 for data storage to store pluralities of the first electronic signal, the second electronic signal, and the third electronic signal.
  • a signal preprocessor 144 can optionally be provided to preprocess the electrical signals received from brain signal sensor 110, muscular signal sensor 120, and movement sensor 130.
  • Processor 140 computes with a plurality of signal processing algorithms, either available in public domain (machine learning, pattern recognition, neural -networks, adaptive control, filtering, online optimization and many other well-known or yet-to-be-developed techniques) or customized by the designer to accommodate desired operation of system 100.
  • Processor 140 also implements training/adaptation/customization algorithms in order to adjust for the expected variations in the required alarm threshold and proper activation of the safety device for each user.
  • a safety device 150 is electronically coupled to processor 140 such that, when the result meets a predetermined threshold value as determined by the software/hardware-based algorithms, safety device 150 is activated.
  • safety device 150 can be a deployable airbag 152 or a plurality of airbags 152.
  • Airbag 152 can be worn about the user's waist, neck/collar, and/or ankles and can be secured to the user by a releasable securing mechanism, such as, for example, a hook and loop attachment.
  • An exemplary airbag that can be used with the present invention is disclosed in U.S. Patent No. 7,017, 195 to Bachman et al., which is incorporated herein by reference.
  • Safety device 150 can be a disposable, for one time use or a multiple usage system.
  • safety device 150 can be another type of safety device, such as, for example, an exoskeleton 160, shown FIG. 4. Exoskeleton 160 can be worn by the user and can be activated to become rigid to prevent the user from falling upon determination by processor 140 that the user has imminent likelihood of falling. Safety device 150 can be disposable, for one time use or a multiple usage system.
  • System 100 can be triggered to activate when the user experiences different sensations. For example, system 100 can activate when the user senses an impending fall and actually begins to fall. Alternatively, system 100 can activate when the user faints or passes out (having no sense of an impending fall), and actually begins to fall. It can also detect changes in the motion, gait and postural stability patterns over time.
  • system 100 Prior to a user initiating use of system 100, it may be desirable for the user to "train" system 100 with regard to typical movements of user that may be associated with particular brain and or muscular signals.
  • user can be connected to system 100 and blindfolded or immersed in a virtual reality system, such as, for example, developed by Oculus VR, LLC, located in Irvine, California. The user can then be provided with the sensation of falling while recording brain activity with brain signal sensor 100, muscle activity with muscular signal sensor 120 and movement with movement sensor 130. This process can be repeated a plurality of times.
  • Processor 140 determines brain and muscular signals associated with the user sensing and reacting to an impending fall and uses the values of such signals to determine threshold values for activating safety device 150.
  • processor 140 also implements training/adaptation/customization algorithms in order to adjust for the expected variations in the required alarm threshold and proper activation of the safety device per each user.
  • brain signal sensor 110 receives a first electrical signal from the brain, indicating a significant increase in brain activity, resulting from the user realizing that he/she is about to fall.
  • Brain signal sensor 110 generates a first electronic signal based on the first electrical signal and transmits the first electronic signal to processor 140.
  • muscular signal sensor 120 receives a second electrical signal from the user's musculature, indicating a sudden contraction of muscles to brace the user for the impending fall, such signal may be received for example, within the range of between about 100 to about 800 ms. If such muscular activity is present, muscular signal sensor 120 generates a second electronic signal based on the second electrical signal and transmits the second electronic signal to processor 140. If such a movement is experienced, movement sensor 130 receives a movement sensation from the user, indicating the likely start of the fall. Movement sensor 130 generates a third electronic signal based on the movement sensation and transmits the third electronic signal to processor 140.
  • Processor 140 processes the first electronic signal, the second electronic signal, and the third electronic signal to generate a result and transmits an activation signal to safety device 150 if the data as processed by the algorithms warrants that the result meets an activation decision value.
  • the activation signal activates safety device 150 to mitigate damage and/or injury to the user when the user falls.
  • system 100 can activate safety device 150 based on a different set of parameters.
  • brain signal sensor 110 receives a first electrical signal from the brain, indicating a significant decrease in brain activity, resulting from the user losing consciousness.
  • Brain signal sensor 110 generates a first electronic signal based on the first electrical signal and transmits the first electronic signal to processor 140.
  • muscular signal sensor 120 receives a second electrical signal from the user's musculature, indicating a sudden change of activation of muscles, also due to loss of consciousness. Muscular signal sensor 120 generates a second electronic signal based on the second electrical signal and transmits the second electronic signal to processor 140.
  • movement sensor 130 receives a movement sensation from the user, indicating the start of the fall. Movement sensor 130 generates a third electronic signal based on the movement sensation and transmits the third electronic signal to processor 140.
  • Processor 140 processes the first electronic signal, the second electronic signal, and the third electronic signal to generate a result and transmits an activation signal to safety device 150 if the data as processed by the algorithms warrants that the result meets a threshold value.
  • the activation signal activates safety device 150 to mitigate damage and/or injury to the user when the user falls.
  • processor 140 interprets those signals as the user realizing that he/she is about to fall and waits for movement sensor 130 to determine that the user is actually falling before transmitting the signal to activate safety device 150.
  • processor, 140 interprets those signals as the user fainting and waits for movement sensor 130 to determine that the user is actually falling before transmitting the signal to activate safety device 150.
  • system 100 is described above as requiring activation of brain signal sensor 110, muscular signal sensor 120, and movement sensor 130 in a specific order in order to activate safety device 150
  • the present invention also contemplates a multiplicity of activation decision patterns and algorithms which are based on signals from the brain signal sensor 110, muscular signal sensor 120, and movement sensor 130 in a variety of methods including different time sequential order, as well as the activation of only one or two of brain signal sensor 110, muscular signal sensor 120, and movement sensor 130 in order to activate safety device 150.
  • System 100 also has the ability to "learn" the user and recalibrate the threshold values for activation of safety device 150 based on the user and the user's daily activities.
  • the threshold value of the first electronic signal may be raised.
  • the activation decision parameters such as the threshold values for either/both the second and third electronic signals, respectively, can be recalibrated if the remaining electronic signals do not indicate an impending fall.
  • a flowchart 400 describing an exemplary arrangement and operation implementing the "learning'Vadaptation of system to a specific user is shown in FIG. 5.
  • Module 141 provides the currently operational fall detection algorithm and parameters;
  • module 142 is a learning adaptation algorithm;
  • module 143 is an actual activation signal;
  • module 144 is a tuning and parameter adjustment signal;
  • module 145 is the access to historical measurements data for learning;
  • module 146 is data acquisition and storage.
  • Module 142 constantly monitors correlation between the actual activation 143 and the sensory data in 146 received by brain sensor 110, muscular sensor 120, and motion sensor 130 and adjusts the parameters in 141 whenever a "mismatch" is present, meaning that, whenever activation signal 141 activates safety device 150 but no falls occurred, or alternatively, when a fall occurs with safety device 150 not being activated.
  • Common gradient descent, least square estimation, adaptive control, machine learning and other known adaptive signal processing techniques may be employed for such adjusting.
  • FIGS. 6-9 correspond to exemplary embodiments of the invention using one or more sensors including a vestibular (and startling/surprise related) motion sensor.
  • FIG. 6 illustrates a schematic drawing of an event detection system using a vestibular (and startling/surprise related) motion sensor according to an exemplary embodiment of the present invention.
  • the event detection system may include at least one sensor is configured to obtain one or more signals from a user.
  • the at least one sensor may include a vestibular (and startling/surprise related) motion sensor 210.
  • the vestibular (and startling/surprise related) motion sensor 210 can be specially configured to include dedicated electroencephalogram (EEG), electromyography (EMG), electrooculography (EOG) and/or near infrared spectroscopy (MRS) components configured to obtain signals from the user which are indicative of the state of the user's vestibular (and startling/surprise related) system.
  • the specially configured and dedicated sensors may be specially configured to obtain vestibular (and startling/surprise related) signals.
  • Vestibular (and startling/surprise related) signals include any signal pattern obtained from the user which indicates the state of the user's vestibular (and startling/surprise related) system.
  • Specialized EEG, EMG, EOG, NIR sensors and the like may obtain and extract vestibular and startling/surprise related signals within 200-300 milliseconds of the onset of body motion.
  • the specialized EEG, EMG, EOG, NIRS sensors and the like may be specially configured to improve signal to noise ratios.
  • the vestibular and startling/surprise related sensors may be located at the vicinity of the vestibular nerves (i.e., behind the ears), near the Vestibular Occular Reflex (VOR) zone (i.e., near eye muscles) and on the skin in places associated with the main muscle groups involved in the Vestibular Spine Reflex (VSR). Such areas which are not covered with hair and are smooth skin also enable cleaner signals with patch like sensors.
  • the vestibular (and startling/surprise related) system includes central projections which participant in three major classes of reflexes: (1) maintaining equilibrium and gaze during movement, (2) maintaining posture, and (3) maintaining muscle tone.
  • observing the vestibular (and startling/surprise related) system may provide an indication of whether a movement is intentional or unintentional.
  • the event detection system provides means for detecting events such as unintentional movement (i.e., fall or startling movement).
  • unintentional movement i.e., fall or startling movement.
  • the vesibular afferent signals are generated in both active (planned) and passive (unplanned/surprising) movements while the efferent signals descending to activate muscles groups in VOR and VSR are suppressed by top-down signals from other cortical areas anticipating such motion and preventing the reflexive action.
  • the first of the reflexes mediated by the vestibular (and startling/surprise related) system is the vestibule-ocular reflex (VOR).
  • VOR helps coordinate the head and eye movements necessary to keep a user's gaze fixated on objects of interest during movements of the head. For example, consider the horizontal movement of the eyes to the right. This movement requires contraction of the left medial and right lateral rectus muscles.
  • Vestibular (and startling/surprise related) nerve fibers originating in the left horizontal semicircular canal project to the medial and lateral vestibular (and startling/surprise related) nuclei.
  • Excitatory fibers from the medial vestibular (and startling/surprise related) nucleus cross to the contralateral abducens nucleus, which has two outputs.
  • One of these outputs is a motor pathway that causes the lateral rectus of the right eye to contract; the other output is an excitatory projection that crosses the midline and ascends via the medial longitudinal fasciculus to the left oculomotor nucleus, where it activates neurons that cause the medial rectus of the left eye to contract.
  • inhibitory neurons project from the medial vestibular nucleus to the left abducens nucleus, directly causing the motor drive on the lateral rectus of the left eye to decrease and also indirectly causing the right medial rectus to relax. Sensing activation of the VOR by sensing any of the outputs, excitations and inhibitions discussed above will aid in the early detection of motion.
  • the second of the reflexes mediated by the vestibular system is essential for postural adjustments of the head and includes the vestibulo-collic reflex (VCR).
  • VCR is a mechanism for controlling neck muscles to control for the head's orientation.
  • the anatomical substrate for the VCR involves the medial vestibular nucleus. Axons from the medial vestibular nucleus descend in the medial longitudinal fasciculus to reach the upper cervical levels of the spinal cord. This pathway regulates head position by reflex activity of neck muscles.
  • the pathway is activated in response to stimulation of the semicircular canals from rotational accelerations of the head. Accordingly, sensing the activation of the VCR can aid in the early detection of motion.
  • the third of theses reflexes mediated by the vestibular system is the vestibulo-spinal reflex (VSR) of the body.
  • the VSR is a mechanism for altering the muscle tone, extension and position of the limbs and head with the goal of supporting posture and maintaining balance of the body and head.
  • the anatomical substrate for the VSR for the body is mediated by a combination of pathways, including the lateral and medial vestibulospinal tracts and the reticulospinal tract.
  • the inputs from the otolith organs project mainly to the lateral vestibular nucleus, which in turn sends axons in the lateral vestibulospinal tract to the spinal cord.
  • VSR is an assemblage of several reflexes named according to the timing (dynamic vs. static or tonic) and sensory input (canal, otolith or both). While terminology varies among authors, the term VSR usually also implies motor output to skeletal muscle below the neck, or in other words, it excludes the neck reflex which is called the VCR (discussed above).
  • extensor tone when the body is pitched, extensor tone changes according to the position of the head with respect to horizontal. Extensor tone is maximal when the angle of the head is 45 degrees with respect to horizontal (i.e. head is nose up as well as an additional 45 degrees towards upright). Extensor tone is minimal when the head is nose-down and pointing an additional 45 degrees down. There is also a "righting reflex”. When the position of the head or body changes, reflex movements occur that tend to return the head or body to the normal posture. Accordingly, sensing activation of the VSR will therefore aid in the early detection of motion.
  • the at least one sensor 212 may include a vestibular (and startling/surprise related) motion sensor 210 having one or more specialized EEG, EMG, EOG, MRS sensors and the like configured to obtain vestibular (and startling/surprise related) signals, including vestibular (and startling/surprise related) signals indicative of the three reflexes discussed above (i.e., the VOR, the VCR and the VSR).
  • a vestibular (and startling/surprise related) motion sensor having one or more specialized EEG, EMG, EOG, MRS sensors and the like configured to obtain vestibular (and startling/surprise related) signals, including vestibular (and startling/surprise related) signals indicative of the three reflexes discussed above (i.e., the VOR, the VCR and the VSR).
  • Each of the dedicated EEG, EMG, EOG, and/or RS sensors of the vestibular (and startling/surprise related) motion sensor may be specially configured to obtain vesti
  • the sensors may be specially configured so that they can obtain the vestibular (and startling/surprise related) signals such as those present in the vestibulocochlear nerve (cranial nerve VIII). Obtaining vestibular (and startling/surprise related) signals from the vestibulocochlear nerve is difficult as the vestibulocochlear nerve is located deep within the skull.
  • the dedicated EEG, EMG, EOG, and/or MRS sensors of the specially configured vestibular (and startling/ surprise related) sensor may be specially configured and positioned so that they may obtain vestibular (and startling/surprise related) signals.
  • the dedicated vestibular (and startling/surprise related) may be placed in the vicinity of the vestibular nerves (i.e., behind the ears), near the Vestibular Occular Reflex (VOR) zone (i.e., near eye muscles) and on the skin close to main muscle groups associated with Vestibular Spine Reflex (VSR). Areas that are not covered with hair and are smooth skin may be preferred so that cleaner signals are obtained.
  • the dedicated EEG sensors may include high-density mobile EEG sensors miniaturized for optimal placement of an electrode array to obtain vestibular (and startling/surprise related) signals.
  • the EEG sensors can be placed into glasses, a hearing aid, and the like.
  • the volitional muscle activity may be obtained by the dedicated EMG sensors of the vestibular (and startling/surprise related) sensor that are configured to record from surface electrodes placed over the tonically-activated sternocleidomastoid (SCM) muscles.
  • the dedicated EMG sensors may include a noninverting surface electrode which is placed at the middle third of the sternocleidomastoid muscles, an inverting electrode placed at the sternoclavicular junction, and a ground electrode placed on the forehead.
  • Vestibular (and startling/surprise related) signals can also be recorded from the extraocular muscles using surface electrodes placed near (approximately inferior to) the eyes.
  • the surface electrodes may obtain vestibular (and startling/surprise related) signals including the ocular vestibular (and startling/surprise related) evoked myogenic potential (oVEMP) which is a manifestation of the VOR.
  • the vestibular (and startling/surprise related) signal obtained by this method is particularly strong during up-gaze when the inferior oblique muscle is activated.
  • the oVEMP can be obtained by eye-tracking by one or more cameras and/or electrooculography (EOG).
  • the vestibular (and startling/surprise related) motion sensor 210 can be configured to sense volitional muscle activity data (VEMG) activity including vestibular (and startling/surprise related) evoked myogenic potential (VEMP).
  • VEMG volitional muscle activity data
  • VEMP evoked myogenic potential
  • the system of FIG. 6 may also include one or more additional sensors such as an accelerometer 220A, gyroscope 220B, an EEG sensor 220C, an muscular sensor 220D, an EOG sensor 220E, a near-infrared spectroscopy (MRS) sensory 220F, and an eye-tracking camera 220G.
  • the sensors 220A-220G may obtain signals indicative of brain and muscle activity which are not limited to vestibular (and startling/surprise related) signals.
  • the one or more sensors may include an accelerometer 220A.
  • the accelerometer 220A may be a micro-machined piezoresistive accelerometer similar to those currently used in various industrial applications.
  • the configuration of the cantilever structures in piezoresistive accelerometers is similar to those in capacitive accelerometers, while their electrical measuring mechanisms are different.
  • piezoresistive accelerometers a piezoresistor is often patterned on a thin suspending cantilever which connects the proof mass and the supporting frame. Due to the mechanical flexibility of the cantilever, a large mechanical strain occurs as the external acceleration displaces the proof mass. The strain is derived from the electrical resistance change in the piezoresistor.
  • Piezoresistive accelerometers can be fabricated by both surface micromachining and bulk micromachining. By using a piezoresistor as the sensing component, this type of accelerometers is advantageous due to the relatively simple configuration and fabrication. However, piezoresistive accelerometers are highly vulnerable to the temperature variation. Improved designs include the use of a large proof mass, integration with a temperature compensation circuitry, and the monolithic implementation with CMOS electronics.
  • Piezoelectric accelerometers have similar configuration to their piezoresistive counterparts, but measure the acceleration from the electrical voltage induced by the mechanical displacement of the cantilever. A notable difference is that piezoelectric accelerometers only respond to dynamic signals while the piezoresistive sensors can measure displacements under low and zero frequencies. Either type of accelerometer may be used in connection with an exemplary embodiment of the device.
  • the accelerometer 220A may include tunneling accelerometers.
  • Tunneling accelerometers take advantage of the phenomenon that occurs when a conductive sharp tip and a counter electrode are positioned at a small gap distance on the order of 10 A.
  • Such accelerometers may be miniaturized for use in measuring vestibular (and startling/surprise related) signals such as VEMP.
  • the tunneling accelerators can be mounted in a hearing aid (see FIGS. 7 A and 7B), in glasses (see FIG. 8), in an implantable sensor, on a safety device (see FIG. 9), on an external surface of the body (e.g., the skin), or the like.
  • the accelerometer 220A is implantable and the temperature resistance of the piezoelectric accelerometers may provide additional movement information.
  • the one or more sensors may also include a gyroscope 220B, a sensor configured to measure the rotary rate of an object.
  • the gyroscope 220B can include micro-gyroscopes which utilize the Coriolis effect to convert the rotary motion of the subject into a measurable linear motion.
  • the Coriolis effect refers to the generation of an imaginary force (Coriolis force) perpendicular to the moving direction of the subject within a rotating coordinate system.
  • the rotary rate can therefore be determined using the above described sensing mechanisms of measuring linear accelerations.
  • the gyroscope 220B sensor is especially well- suited to use in the embodiments of the system illustrated in FIGS. 7 A and 8B which include devices resembling a hearing aid.
  • the one or more sensors may also include an EEG sensor 220C.
  • EEG systems to produce electrooculographic (and electronystagmography) data indicative of motion information including vestibular (and startling/surprise related) functioning has been described (see, e.g., Toglia, Clin. Electroencephalogr, Oct; 19(4):225-30 (1988)), the contents of which are incorporated herein by reference).
  • Toglia Clin. Electroencephalogr, Oct; 19(4):225-30 (1988)
  • EEG data acquisition was demonstrated in a moving patient, and then only in persons moving linearly (see, e.g., Nolan, et al, Proceedings of the 4th International IEEE EMBS Conference on Neural Engineering, Antalya, Turkey, April 29 - May 2, 2009).
  • the vestibular (and startling/ surprise related) motion sensors 210 may include specially configured high-density mobile EEG sensors configured to obtain, process and transmit one or more vestibular (and startling/ surprise related) signals
  • other EEG sensors 220C may be used in addition to the EEG sensors specially configured for obtaining vestibular (and startling/ surprise related) signals.
  • the EEG sensors 220C can include one or more leads in contact with the user's scalp.
  • the one or more sensors may also include a muscular sensor 220D.
  • the muscular sensor 220D may be adapted to receive or obtain an electrical signal indicative of muscle activity.
  • the muscular sensor 220D can include a processor configured to convert the obtained or received electrical signal into an electronic signal. Alternatively the conversion from an electrical signal to an electronic signal can occur at pre-processing 244 or by the one or more processors 240.
  • the muscular sensor 220D can transmit the electrical (or electronic) signal to the one or more processors 240.
  • the muscular sensor 220D can include an electromyography ("EMG”) sensor.
  • EMG electromyography
  • the muscular sensor 220D can be attached to the user's thighs, picking up sensory information from quadriceps or hamstrings; upper arms; neck; or other major muscle groups whose reflexive electrical activation is being pursued and monitored.
  • the muscular sensor 220D can be utilized through connection (implanted, partially implanted, or on the skin) to one or more major muscle masses in the body.
  • Such a muscular sensor 220D can be employed to provide one or more electrical signals configured to confirm motion detected via the vestibular (and startling/surprise related) motion sensor 210.
  • the one or more sensors may also include a sensor configured for electrooculagraphy (EOG).
  • EOG sensor 220E may be configured to record electrical signals proximate the eyes.
  • the electrooculogram obtained by the EOG sensor may be used to determine eye movement which may be indicative of body movement.
  • the one or more sensors may also include a near-infrared spectroscopy (MRS) sensor 220F.
  • the MRS sensors 220F may include wearable modules which are configured to communicate with a smartphone or tablet as a receiver such as those made by Alps Electronics, Ltd.
  • a fully wireless MRS device is also described by Achigui, et al, Microelectronics Journal, 39(10):Pages 1209-1217 (2008), incorporated herein by reference.
  • the MRS sensor 220F makes use of dynamic threshold transistors (DTMOS) for low voltage (IV), low power and low noise enhancement.
  • DTMOS dynamic threshold transistors
  • the design is composed of a transimpedance amplifier (TIA) and an operational transconductance amplifier (OTA).
  • the OTA differential input pairs use DTMOS devices for input common mode range enhancement.
  • the OTA is fabricated in a standard 0.18 ⁇ CMOS process technology. Measurements under a 5 pF capacitive load for the OTA gives a DC open loop gain of 67 dB, unity frequency gain bandwidth of 400 kHz, input and output swings of 0.58 and 0.7 V, a power consumption of 18 ⁇ , and an input referred noise of 134 nV/VHz at 1 kHz without any extra noise reduction techniques.
  • the one or more sensors of the device may also include an eye-tracking camera 220G.
  • the eye-tracking camera 220G may be configured to track eye movements indicative of intentional and unintentional movements.
  • Each of the one or more sensors including the specially configured vestibular (and startling/surprise related) motion sensor 210 and the remaining sensors (accelerometer 220A, gyroscope 220B, an EEG sensor 220C, an muscular sensor 220D, an EOG sensor 220E, a near-infrared spectroscopy (NIRS) sensory 220F, and an eye-tracking camera 220G, discussed above) may be in communication with each other.
  • the one or more sensors may operate independently of each other.
  • FIG. 5 illustrates and embodiment having seven sensors, any combination of sensors, including those with fewer or greater number of sensors may be used.
  • the system 200 can also include one or more sensors 220A-220G configured to detect movement and adapted to receive or obtain an electrical signal indicative of movement.
  • the one or more sensors 220A-220G can include a processor configured to convert the obtained or received electrical signal into an electronic signal. Alternatively, the conversion from an electrical signal to an electronic signal can occur at pre-processing 244 or by the one or more processors 240.
  • the one or more sensors 220A-220G can transmit the electrical (or electronic) signal to the one or more processors 240.
  • the one or more sensors can include one or more of an accelerometer, a gyroscope and other known or as yet unknown velocity, displacement, acceleration, jerk or other movement sensors.
  • the one or more sensors can be attached to the user's chest, or may also be incorporated into a hearing aid (see FIGS. 6A and 6B), set of glasses (see FIG. 7), or implantable sensor in a hermetically sealed housing.
  • Other sensors may be located anywhere on the user's core and head, depending on the product design, convenience, and other potential user disabilities.
  • these sensors may also be incorporated in the surrounding structure around the person such as a car, a room, a clinic, a hospital, a bike, a ski set, etc.
  • the sensors may be non-contact sensors which use cameras, optics, infrared, wide wavelengths, ultrasound and other types of radarlike methods to obtain signals from the brain, central nervous system, peripheral nervous system, and the body.
  • Each of the one or more sensors may obtain one or more signals from a user.
  • the signals obtained by the vestibular (and startling/surprise related) motion sensor 210 may include vestibular (and startling/surprise related) signals indicative of the state of the user's vestibular (and startling/surprise related) system.
  • the vestibular (and startling/surprise related) signals may include signals propagating along the direct pathway from the vestibular (and startling/surprise related) afferents (i.e., the horizontal and vertical semicircular canals, utricular afferents, saccule fibers, etc.) to the vestibular nuclei.
  • the signals may be present within approximately about 100 ms of the onset of unintentional motion activity. More particularly, the vestibulo-spinal reflex (VSR) may be present within approximately about 70 ms of the onset of unintentional motion activity, and the vestibulo-ocular reflex (VOR) may be observed within approximately about 60 ms of the onset of unintentional motion activity.
  • VSR vestibulo-spinal reflex
  • VOR vestibulo-ocular reflex
  • non-vestibular signals which propagate along the indirect pathways may not be present until over approximately about 150 ms of the onset of unintentional motion activity.
  • vestibular (and startling/surprise related) signals allow for the earlier detection of unintentional motion activity.
  • the vestibular (and startling/ surprise related) signals may be suppressed when motion activity is intentional (such as when a person laughs, jumps, walks, etc.) and be present when motion activity is unintentional (such as when a person falls, is startled, etc.).
  • the vestibular (and startling/surprise related) signals may be indicative of events.
  • Vestibular (and startling/surprise related) signals may also be encoded by the thalamus and cortex.
  • alternative embodiments of the device may obtain vestibular (and startling/surprise related) signals from vestibular (and startling/surprise related) motion sensors 210 positioned about the thalamus and cortex.
  • the signals obtained from the remaining sensors 220A-220G may also be indicative of the user's vestibular (and startling/surprise related) and non-vestibular (and startling/surprise related) systems.
  • the electrical signals obtained by each of the sensors can be processed into one or more electronic signals by one or more processors.
  • the one or more processors can be located at each respective sensor 210, 220A-G, or at a common processing element 240.
  • the one or more processors 240 can be organized in a processing block 235 further comprising a data repository 242 and/or a signal preprocessor 244.
  • One or more of the electronic signals can be transmitted by each of the sensors 210, 220A-G to a pre-processing unit 244.
  • the preprocessing unit 244 may be configured to apply signal processing techniques which enable the electronic signals to be used by the processing element 240.
  • the signal processing techniques may include filtering, conditioning, outlier and artifact removal, amplification and the like.
  • the electrical or electronic signals can be transmitted directly to the processing element 240.
  • the transmission of electrical and/or electronic signals can be wired or wireless.
  • the processing element 240 may include one or more processors coupled to the at least one sensor 210, 220A-G, either directly or via the pre-processing unit 244.
  • the processing element 240 may also be coupled to a memory 241 configured to store computer readable instructions.
  • the computer readable instructions stored on the memory 241 may cause the processing element 240 to receive the one or more signals from the at least one sensor and extract vestibular (and startling/surprise related) data from the one or more signals.
  • one or more processors 240 can be electronically coupled to each of the one or more sensors 210, 220A-220G.
  • the one or more processors 240 can be electronically coupled to the sensory device unit 212 comprising each of the one or more sensors 210, 220A-220G.
  • a single sensory device unit 212 may package the one or more sensors into a single unit that can be attached to a single location on the user. An exemplary location for the sensory device unit 212 can be on the back of the user's neck.
  • the vestibular (and startling/surprise related) motion sensor 210 can obtain the vestibular (and startling/ surprise related) signal from neurological signals generated by the vestibular nuclei which are located on either side of the brain stem.
  • the muscular sensor 220D can be used to sense movement of the trapezius and sternocleidomastoid muscles of the neck.
  • the one or more processors 240 can include an electronic microprocessor and can be powered by an electrical power source, such as a battery (not shown).
  • the one or more processors can include circuit-based processes, including possible implementation as a single integrated circuit (such as an ASIC or FPGA), a multi-chip module, a single card, or a multi- card circuit pack.
  • various functions of circuit elements may also be implemented as processing blocks in a software program. Such software may be employed in, for example, a digital signal processor, micro-controller, or general-purpose computer.
  • the processing element 240 may extract vestibular (and startling/surprise related) data from the one or more signals. Extracting vestibular (and startling/surprise related) data may include the steps of signal conditioning, noise reduction, normalization, and outliers and artifacts removal. In one embodiment, extracting vestibular (and startling/surprise related) data may include a time window limitation. The time window limitation may be based on an understanding of the vestibular system's anatomy and physiology. In one embodiment the time window limitation may be between about 0 and 300 milliseconds.
  • the processing element 240 may include software which provides a system for real time extraction of several features from the collected data.
  • the features that may be extracted in real-time from the collected data may be defined and selected off-line based on one or more pilot studies used to generate the predetermined data.
  • the features extracted from the one or more signals may include temporal and frequency distribution moments, dynamic ranges, short and long term cross- correlations, and the like. These features may comprise the vestibular (and startling/surprise related) data.
  • These features may be stored as vectors in one or more data sets. Each data sets may incorporate features from all the sensors. Clustering and advanced machine learning methods, including deep learning techniques (see,
  • the computer readable instructions stored on the memory 241 may also cause the processing element 240 to determine whether the user is undergoing an event based on the extracted vestibular (and startling/surprise related) data from the one or more signals.
  • determining whether the user is undergoing an event may also include extracting non-vestibular (and startling/surprise related) data from the one or more signals and using both the vestibular (and startling/surprise related) and non-vestibular (and startling/surprise related) data.
  • the vestibular (and startling/surprise related) data extracted from the one or more signals is compared to predetermined data in order to determine whether the user is undergoing an event. Determining whether the user is undergoing an event such as a fall or a startling movement may involve comparing vestibular (and startling/surprise related) data extracted from the obtained one or more signals to predetermined data. Determining whether the user is undergoing an event may also involve comparing non-vestibular data (such as EEG/EMG/EOG signals that are non-indicative of the vestibular (and startling/surprise related) system, gyroscope, and acceleration profiles) with predetermined data.
  • non-vestibular data such as EEG/EMG/EOG signals that are non-indicative of the vestibular (and startling/surprise related) system, gyroscope, and acceleration profiles
  • the predetermined data may include population data and user data.
  • the population data may be indicative of the state of each user in a population during an event or non-event.
  • the population data may include data sets which describe the profile of signals obtained from the one or more sensors from multiple users in event or non-event states.
  • the population data may be stored on the memory 241 coupled to the processor 240. Alternatively, the population data may be stored remotely from the device and accessed by the processor via a wireless internet connection or the like.
  • the predetermined data may also include user data.
  • the user data may be indicative of the state of the user's vestibular (and startling/surprise related) system during previous event or non-event states.
  • the user data may be obtained during a trial period where the user undergoes push and jump trials.
  • the push and jump trials the user's vestibular (and startling/surprise related) system activity before, during, and after push and jump events are recorded to determine the response of the user's vestibular (and startling/surprise related) system to intentional movements such as a jump and unintentional movements such as a push.
  • Preliminary movement studies have illustrated suppression of vestibular (and startling/surprise related) activity in intentional movements.
  • Preliminary movement studies have also illustrated a lack of suppression of vestibular (and startling/surprise related) activity in unintentional movements.
  • the predetermined data and user data may also include non-vestibular data such as acceleration and gyroscope profiles.
  • non-vestibular data such as acceleration and gyroscope profiles may be compared against predetermined data comprising non-vestibular data from the population and user in order to make a determination as to whether the user is undergoing an event such as a fall or a startling movement.
  • the predetermined data can be used in connection with one or more classification algorithms, so that data and information extracted from the one or more signals obtained by the device can be classified as an event or non-event.
  • the classification algorithms may be based in machine learning, pattern recognition, and clustering models.
  • a simple thresholding algorithm may be applied where the predetermined data is used to establish thresholds which characterize motion information as indicative of a non-event or event.
  • the vestibular (and startling/surprise related) and/or non-vestibular (and startling/surprise related) signals processed by the one or more processors 240 may be compared to the thresholds to determine whether the user is undergoing an event or non-event.
  • the determination of whether a user is undergoing an event may be done in substantially real-time.
  • the system may be configured to provide continuous monitoring to a user.
  • the processor can make an accurate determination as to whether a movement is an event within about 50 to 200 miliseconds and transmit the signal to activate the safety device.
  • the one or more processors 240 determines that a user is undergoing an event by comparing the vestibular (and startling/surprise related) data to predetermined data one or more activation signals can be transmitted from the one or more processors 240 to a safety device 250. Upon receiving the one or more activation signals, the safety device 250 can deploy.
  • the one or more processors 240 can include software code including instructions to determine whether the user is undergoing an event, generate and transmit the activation signals, and the like.
  • the one or more processors 240 can optionally generate and transmit alarm activation signals configured to activate an alarm system 255 coupled to the one or more processors 240.
  • the alarm system can be configured to provide an alert to the user, a caregiver, a medical professional and the like.
  • the alert can include a communication such as a text message, phone call or e-mail, a loud beeping noise, a light that is triggered at a nurses' station, and the like.
  • the alarm activation signals can be transmitted from the one or more processors 240 to the alarm system 255 in a wired or wireless method.
  • the one or more processors 240 can optionally transmit data and information relating to the electrical and electronic signals to one or more data storage devices or data repositories 242.
  • the one or more processors 240 can be communicatively coupled to a data repository 242 including memory.
  • the data repository 242 can be located with the one or more processors in the processing block 235 or the data repository 242 can be located away from the one or more processors 240 (as illustrated in FIG. 6).
  • the communicative coupling may be wired or wireless.
  • the data repository 242 can be configured to collect and store the data and information including raw signals from each of the sensors 210, 220, 230 and results from the one or more processors 240.
  • the data and information can be archived, analyzed, and communicated to third parties in order to gather data and information regarding medical conditions.
  • safety device 250 can be optionally omitted.
  • data received and stored by the one or more processors 240 can be transmitted to a third party device, such as that belonging to a physician, for an epilepsy patient who is using system 200.
  • the data received and stored by the third party device can be used to interpret whether the patient's motions are unusually rigid, perform gait analysis and/or postural stability analysis.
  • the data and information can be used to determine whether a patient's medication dosage needs to be adjusted. For example, if the data indicates that the patient exhibits volatility and fluctuations in motor behavior over a period of time, a patient's medication dose may be adjusted. In this manner, at least one embodiment of the system can provide continuous patient monitoring.
  • the physician can contact the patient and tell the patient to adjust the dosage of medication.
  • the patient can be informed directly, such as, for example, through a patient-owned/operated electronic device through which the data indicates improper dosing and that the patient should either adjust the dosing or contact the patient's physician to inquire into changing the dosing prior to doing so.
  • the exemplary medical condition discussed above is epilepsy, those skilled in the art will recognize that system 200 can be used to indicate aberrations in the motions of a patient with other physical issues/ailments including, but not limited to, chronic dizziness and other maladies.
  • data and information from a single user can be stripped of all personally identifying information either prior to transmission to the data repository 242 or upon receipt at the data repository 242.
  • the data repository 242 can store the data and information from multiple users of the system 200.
  • the data and information can be combined to form a database for further clinical studies to better understand an illness, disease state, the effects of a pharmaceutical drug or therapy, and the like.
  • the data may also be uploaded to a web service or the like.
  • the combined data and information from multiple users may form, at least in part, the population data used by the device to determine whether a user is undergoing an event.
  • a safety device 250 can be electronically coupled to the one or more processors 240 such that when it is determined that the user is undergoing an event by comparing the vestibular (and startling/surprise related) data to predetermined data an activation signal can be transmitted to the safety device 250.
  • Safety devices can include airbags (see above with regards to FIG. 3), exoskeletons (see above with regards to FIG. 4), hip protectors (see below with regards to FIG. 9), and the like.
  • Examples of a safety device 250 include the wearable airbags found in U.S. Pat. No. 5,500,952, U.S. Pat. No. 7,017, 195, U.S. Pat. No. 7, 150,048 and US. Patent Application Publication numbered 2005/0067816. These references relate to wearable airbag clothes coupled with a non-vestibular sensor to sense the movement of a person wearing the wearable airbag clothes.
  • the safety device 250 i.e., airbag
  • the airbag cushion can include an inner cavity that is sized and shaped so that during inflation the airbag is quickly filled with gas, but during impact the gas outflow from the airbag is delayed.
  • the inner cavity of the airbag cushion can be sectioned into a plurality of inner sections/chambers.
  • each of the inner sections can be connected to the neighboring sections with one or more air passages.
  • Each of the inner sections can comprise multiple inner chambers so that the smaller chambers can inflate faster and can provide enough protection to body parts during impact.
  • Sectional airbag design is advantageous in that it will delay the air outflow when a user falls on top of the airbag.
  • each of the inner section can be connected to the air movement system via a separate valve.
  • the air movement system can comprise a manifold with a number of ports, each of the ports connected to a separate inner section of the cushion.
  • each of the inner sections can be deployed automatically at the same time.
  • the airbag system can comprise one or more vent-valves to provide slow release of the airbag cushion upon a fall impact.
  • an airbag may comprise a pneumatic airbag system, including a pneumatic sub-system.
  • the pneumatic sub-system may also include a gas canister connected to a gas pressure gage and a gas discharge valve connected to a gas outlet.
  • the gas outlet may be connected to a pneumatic tubule, which is split by means of a manifold, from which a pneumatic tubule goes to each gas intake valve of each one of the airbags.
  • the vestibular (and startling/surprise related) motion sensor 210 of system 200 can be used in addition to or as a substitute for the brain signal sensor 100. Accordingly, similar to the training methods discussed above with regards to system 100, system 200 can also utilize training methods to establish appropriate thresholds for the one or more signals obtained from the vestibular (and startling/surprise related) motion sensors.
  • FIG. 7A illustrates an example of a fall and collision detection and injury mitigation system 300 using a vestibular (and startling/surprise related) motion sensor 303 which is coupled to a muscular sensor 305 and a movement sensor 309 according to an exemplary embodiment of the present invention as discussed with regards to FIG. 6.
  • the sensors 303, 305, and 309 can be embedded within a hearing aid 301 device configured to be positioned behind a user's ear.
  • the vestibular (and startling/surprise related) motion sensor 303 is non-invasive.
  • the vestibular (and startling/surprise related) motion sensor 303 can include EEG electrodes configured to be positioned across the scalp so as to pick up cortical waveforms.
  • the EEG electrodes may also be placed against the skin external to the ear canal. In one embodiment, the EEG electrodes may be placed proximate other muscle groups and/or places on the human body which may be involved in motion.
  • the fall and collision detection and injury mitigation system 300 can optionally include a muscular sensor 305 which includes EMG electrodes.
  • the muscular sensor 305 can be coupled to the fall and collision detection and injury mitigation system 300 at a socket 307.
  • the fall and collision detection and injury mitigation system 300 can also optionally include a movement sensor 309 further including one or more accelerometers.
  • the movement sensors 309 may be configured to extract signals which propagate due to the physical transduction phenomena (relative acceleration, velocity, jerk, pressure, voltage etc.). Muscular sensors may exploit the cellular action potentials which propagate along the skin's surface.
  • FIG. 7B illustrates an alternative example of a fall and collision detection and injury mitigation system 400 also using a vestibular (and startling/surprise related) motion sensor 403 which is coupled to a muscular sensor 405 and a movement sensor 409 according to an exemplary embodiment of the present invention as discussed with regards to FIG. 6.
  • the hearing aid-like device 401 can be inserted in direction 411 into a user's ear canal 413.
  • the vestibular (and startling/surprise related) motion sensor 403 is partially invasive.
  • the vestibular (and startling/surprise related) motion sensor 403 further includes EEG electrodes which are positioned within the ear canal and can obtain vestibular signals.
  • the fall and collision detection and injury mitigation system 400 can optionally include a muscular sensor 405 which includes EMG electrodes.
  • the muscular sensor 405 can be coupled to the fall and collision detection and injury mitigation system 400 at a socket 407.
  • the fall and collision detection and injury mitigation system 400 can also optionally include a movement sensor 409 further including one or more accelerometers.
  • FIG. 8 illustrates an alternative example of a fall and collision detection and injury mitigation system 500 also using a vestibular (and startling/surprise related) motion sensor 503 which is coupled to a muscular sensor 505 and a movement sensor 509 according to an exemplary embodiment of the present invention as discussed with regards to FIG. 6.
  • the glasses-like device 501 can rest on the bridge of a user's nose and ears.
  • the vestibular (and startling/surprise related) motion sensor 503 is non-invasive.
  • the vestibular (and startling/surprise related) motion sensor 503 can include eye-tracking cameras configured to obtain information regarding the VOR.
  • the vestibular (and startling/ surprise related) motion sensor 503 can also include one or more electrodes configured to receive EEG 503B and/or EMG 505 signals.
  • the electrodes configured to receive EMG 505 can also receive muscular information.
  • the one or more electrodes configured to receive EEG 503B and/or EMG 505 signals can be positioned on the user's temple.
  • the fall and collision detection and injury mitigation system 500 can also optionally include one or more movement sensors including one or more accelerometers 509A and gyroscopes 509B.
  • the device 501 can transmit data and information wired or wirelessly to a controller 511 including the processing block 235 (discussed above). Alternatively the processing block 235 can be mounted to the frame of the device 501.
  • FIG. 9 shows an exemplary wearable safety device 250 for use with the vestibular (and startling/surprise related) motion sensors 210, 303, 403, 503 according to an exemplary embodiment of the present invention.
  • the safety device 250 is a low profile waist belt.
  • the safety device 250 can be electronically coupled to controller 511 and/or processing block 235 such that when it is determined that data extracted from signals obtained from the user indicate an event the controller 140 and/or processing block 235 can transmit an electronic signal to the safety device 250 in order to activate the safety device 250.
  • the safety device 250 can be configured to protect the hip area, an area known to be prone to injury in elderly patients.
  • the safety device 250 can be camouflaged by or worn under everyday clothing thereby reducing the aesthetically unpleasing factors associated with conventional safety devices.
  • the vestibular (and startling/surprise related) motion sensors also reduce the aesthetically unpleasing factors associated with conventional devices by taking advantage of the latest wearable devices technology and incorporating accepted platforms of eye trackers and hearing aids technology in an ergonomic and user friendly way.
  • user data obtained by experiments was used as predetermined data.
  • the user data was transmitted to and/or stored in a data repository.
  • user data was gathered from experiments using push and jump trials.
  • EEG, EMG, linear and angular acceleration data were obtained from a subject when the subject was pushed or encouraged to fall from a bench at similar acceleration profiles over multiple trials.
  • the push and jump trials utilized a device with 24 bit resolution, 5000 Hz sampling rate, wide band DC-3500 Hz, a large dynamic range of +/- 430 mV, AC and DC measurement modes that were switchable channel by channel, 40-channel amplifier having 32 EEG and 8 bipolar channels.
  • the device also included a 2D accelerometer with a range of +/- 6g and a ID gyroscope.
  • the device obtained vestibular and other signals through EEG electrodes.
  • Passive EEG electrodes were present on a head cap.
  • the electrodes were placed on subject's head in accordance with the international 10-10 system in which electrodes are spaced apart by approximately 10% of the total front-back or right-left distance of the skull.
  • the higher resolution international 10-5 system or the lower resolution international 10-20 system may be used.
  • Any electrode placement that results in obtaining vestibular and other signals may be used.
  • a reference electrode was placed between the left mastoid and the 01 electrode from the international 10-10 system.
  • a ground electrode was placed between the right mastoid and the 02 electrode from the international 10-10 system.
  • the reference electrode may have been attached to one earlobe and the ground electrode may have been attached to the mastoid on the same side of the head.
  • the reference electrode may have attached to one mastoid and the ground electrode may have attached to another.
  • EMG sensors were symmetrically placed about other regions of interest on the subject including under the eye, on the back of the neck, and on the shoulder.
  • a 2D accelerometer and ID gyroscope were placed along the lower back area of the subject.
  • the 2D accelerometer measured movement along the up and down axis parallel to the subject's spine, and the forward and backward movement of the subject.
  • the ID gyroscope aided in measuring the subjects' s rotation about the axis parallel to the subject's spine.
  • a subject stood on the bench with their eyes closed.
  • the subject was connected to a harness, wore a jacket, or other safety equipment.
  • the subject was connected to the device and in particular one or more EEG electrodes, one or more EMG electrodes, one or more accelerometers, and/or one or more gyrometers as described above.
  • the subject was pushed forwards or backwards without notification, causing the subject to fall.
  • the time at which the subject was pushed was recorded.
  • the subject was then asked to stand on the bench with their eyes closed again.
  • the subject was then asked to mimic the fall event.
  • FIG. 10 A composite image from the push and jump trials is depicted in FIG. 10.
  • waveforms from push and jump trials were viewed within a graphical user interface (GUI) 900 of a computer software configured to display data related to the device.
  • GUI graphical user interface
  • Waveforms representative of EEG data 901, waveforms representative of EMG data 903, waveforms representative of acceleration data 905, waveforms representative of gyroscope data 907 were shown over time along a time axis 909.
  • Waveforms associated with a pushing event were found in the region marked as '911 A.
  • Waveforms associated with an intended jumping event were found in the region marked '91 IB.
  • Shortly after a push was initiated 913 a sharp cluster of spikes 915 was observed in waveforms indicative of EEG data.
  • a drastic change in acceleration was observed 917 in the waveforms representative of accelerometer and gyroscope data. The drastic change in acceleration may be attributed to movement, falling, landing, etc.
  • an intentional movement such as a jump 919 there is no sharp cluster of spikes observed prior to the observed drastic change in acceleration 921.
  • the cluster of spikes observed prior to the drastic change in acceleration when the push is initiated may be indicative of vestibular (and startling/surprise related) activity. Accordingly, the push and jump trials indicate that vestibular activity (as illustrated by the cluster of spikes) may be suppressed in intentional motion. Accordingly vestibular (and startling/surprise related) activity may be indicative of falling motion.
  • FIG. 10 shows example data obtained from vestibular (and startling/surprise related) sensors used in connection with an exemplary embodiment of the present invention.
  • the example data indicates a suppression of vestibular (and startling/surprise related) activity in intentional movements.
  • the example data also indicates a lack of suppression of vestibular (and startling/surprise related) activity in unintentional movements.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Veterinary Medicine (AREA)
  • Epidemiology (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Psychology (AREA)
  • Psychiatry (AREA)
  • Biophysics (AREA)
  • Primary Health Care (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Child & Adolescent Psychology (AREA)
  • Developmental Disabilities (AREA)
  • Hospice & Palliative Care (AREA)
  • Social Psychology (AREA)
  • Dentistry (AREA)
  • Physiology (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

Un système permettant de détecter et de répondre à des crises et à des mouvements de l'utilisateur ayant le potentiel de provoquer une blessure.
PCT/US2016/062543 2015-11-18 2016-11-17 Système de détection de crise et d'atténuation des blessures et son procédé d'utilisation Ceased WO2017087680A1 (fr)

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
US201562257220P 2015-11-18 2015-11-18
US62/257,220 2015-11-18
US201615144712A 2016-05-02 2016-05-02
US15/144,712 2016-05-02
US201615208399A 2016-07-12 2016-07-12
US15/208,399 2016-07-12
US15/269,820 2016-09-19
US15/269,820 US9974344B2 (en) 2013-10-25 2016-09-19 Injury mitigation system and method using adaptive fall and collision detection

Publications (1)

Publication Number Publication Date
WO2017087680A1 true WO2017087680A1 (fr) 2017-05-26

Family

ID=58717868

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2016/062543 Ceased WO2017087680A1 (fr) 2015-11-18 2016-11-17 Système de détection de crise et d'atténuation des blessures et son procédé d'utilisation

Country Status (1)

Country Link
WO (1) WO2017087680A1 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111820907A (zh) * 2020-08-28 2020-10-27 江苏无恙半导体科技有限公司 一种便携式癫痫报警装置
FR3108249A1 (fr) * 2020-03-23 2021-09-24 Henri PAYRE Dispositif de protection d’un patient épileptique
CN115530774A (zh) * 2021-06-30 2022-12-30 荣耀终端有限公司 癫痫检测方法和装置
US12224049B2 (en) 2020-12-09 2025-02-11 Eysz, Inc. Systems and methods for monitoring and managing neurological diseases and conditions
WO2025189195A1 (fr) * 2024-03-08 2025-09-12 Rekovar Inc. Dispositif, système et méthode de détection de convulsions néo-natales basés sur l'intelligence artificielle

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070038268A1 (en) * 2005-08-09 2007-02-15 Weinberg Marc S Multi-axis tilt estimation and fall remediation
US20090322540A1 (en) * 2008-06-27 2009-12-31 Richardson Neal T Autonomous fall monitor
US20120197092A1 (en) * 2011-01-28 2012-08-02 Neurosky, Inc. Dry sensor eeg/emg and motion sensing system for seizure detection and monitoring
US20150101112A1 (en) * 2013-10-12 2015-04-16 Joel Abe Balbien Method and Apparatus of Preventing a Fall or Minimizing the Impact of the Fall of an Individual
US20150157252A1 (en) * 2013-12-05 2015-06-11 Cyberonics, Inc. Systems and methods of limb-based accelerometer assessments of neurological disorders

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070038268A1 (en) * 2005-08-09 2007-02-15 Weinberg Marc S Multi-axis tilt estimation and fall remediation
US20090322540A1 (en) * 2008-06-27 2009-12-31 Richardson Neal T Autonomous fall monitor
US20120197092A1 (en) * 2011-01-28 2012-08-02 Neurosky, Inc. Dry sensor eeg/emg and motion sensing system for seizure detection and monitoring
US20150101112A1 (en) * 2013-10-12 2015-04-16 Joel Abe Balbien Method and Apparatus of Preventing a Fall or Minimizing the Impact of the Fall of an Individual
US20150157252A1 (en) * 2013-12-05 2015-06-11 Cyberonics, Inc. Systems and methods of limb-based accelerometer assessments of neurological disorders

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
IBRAHIM, A ET AL.: "Simple fall criteria for MEMS sensors: data analysis and sensor concept.", SENSORS, vol. 14, no. 7, 8 July 2014 (2014-07-08), pages 12149 - 12173, XP055382693 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3108249A1 (fr) * 2020-03-23 2021-09-24 Henri PAYRE Dispositif de protection d’un patient épileptique
CN111820907A (zh) * 2020-08-28 2020-10-27 江苏无恙半导体科技有限公司 一种便携式癫痫报警装置
US12224049B2 (en) 2020-12-09 2025-02-11 Eysz, Inc. Systems and methods for monitoring and managing neurological diseases and conditions
CN115530774A (zh) * 2021-06-30 2022-12-30 荣耀终端有限公司 癫痫检测方法和装置
CN115530774B (zh) * 2021-06-30 2024-03-26 荣耀终端有限公司 癫痫检测方法和装置
WO2025189195A1 (fr) * 2024-03-08 2025-09-12 Rekovar Inc. Dispositif, système et méthode de détection de convulsions néo-natales basés sur l'intelligence artificielle

Similar Documents

Publication Publication Date Title
US9974344B2 (en) Injury mitigation system and method using adaptive fall and collision detection
US12076567B2 (en) System and method for neurostimulation
US11071493B2 (en) Multicomponent brain-based electromagnetic biosignal detection system
KR102100120B1 (ko) 생체 신호를 모니터링하는 방법, 장치 및 컴퓨터 프로그램
US10485471B2 (en) System and method for identifying ictal states in a patient
US12239423B2 (en) Detection of patient conditions using signals sensed on or near the head
US8021299B2 (en) Correlating a non-polysomnographic physiological parameter set with sleep states
US12390115B2 (en) Photoplethysmography based detection of transitions between awake, drowsiness, and sleep phases of a subject
WO2017087680A1 (fr) Système de détection de crise et d'atténuation des blessures et son procédé d'utilisation
KR20170083483A (ko) 실시간 수면장애 감시 장치
WO2017119638A1 (fr) Appareil de surveillance de trouble du sommeil en temps réel
WO2015168151A1 (fr) Système et procédé de suivi de la dynamique du sommeil à l'aide d'informations comportementales et physiologiques
Rajan et al. Analysis of human brain disorders for effectual hippocampus surveillance
CA3050352A1 (fr) Device for monitoring the electro-physiological activity of a subject
CN120241012B (zh) 一种老年心脏患者动态监测系统
Salem et al. Nocturnal epileptic seizures detection using inertial and muscular sensors
Poh Continuous assessment of epileptic seizures with wrist-worn biosensors
Gheryani et al. Epileptic Seizures Detection based on Inertial and Physiological Data from Wireless Body Sensors
Andreoni et al. Example of clinical applications of wearable monitoring systems
Alalaiwi et al. Development of Low Cost “Epi-seizure” Device for the Prediction and Detection of Tonic-Clonic Seizures
Patel et al. Evaluating the feasibility of Fall detection using Single-channel EEG
De Cooman Epileptic Seizure Detection in a Home Environment
WO2025027581A1 (fr) Système et dispositif de surveillance du mouvement des yeux et des paupières
Baga et al. PERFORM: A platform for monitoring and management of chronic neurodegenerative diseases: The Parkinson and Amyotrophic Lateral Sclerosis case
JP2025028636A (ja) 空間共有システムおよび空間共有方法

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: 16867142

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16867142

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 16867142

Country of ref document: EP

Kind code of ref document: A1