US20220351814A1

US20220351814A1 - Stitching related healthcare data together

Info

Publication number: US20220351814A1
Application number: US17/734,955
Authority: US
Inventors: Steven Ryan Facer; Xu Liu
Original assignee: Udo LLC
Current assignee: Ark Group LLC; THE FRASURE FAMILY TRUST
Priority date: 2021-05-03
Filing date: 2022-05-02
Publication date: 2022-11-03
Also published as: US12075189B2; WO2022235635A1; WO2022235626A1; US20240380864A1; US20220353463A1

Abstract

A computing system is configured to access a plurality of healthcare datasets, each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact. The artifact includes at least one of (1) textual data, (2) image data, (3) audio data, or (4) video data associated with a condition of the patient or care of the patient. For each of the plurality of healthcare datasets, the healthcare dataset is labeled with one or more explicit designation labels and one or more contextual substitutable labels. The computing system is further configured to query the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label and/or a same contextual substitutable.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application claims the benefit of (1) U.S. Provisional Application No. 63/183,409, filed May 3, 2021, entitled “Two-Way Camera Operation”, (2) U.S. Provisional Application No. 63/183,411, filed May 3, 2021, entitled “Security and Verification with a Data Sharing Platform”, (3) U.S. Provisional Application No. 63/183,414, filed May 3, 2021, entitled “User Invitation for Accessing Healthcare Data”, (4) U.S. Provisional Application No. 63/183,419, filed May 3, 2021, entitled “Stitching Related Healthcare Data Together”, and (5) U.S. Provisional Application No. 63/183,421, filed May 3, 2021, entitled “Training AI Models for Diagnosing Diseases or Generating Treatment Plans.” All of the aforementioned applications are incorporated by reference herein in their entirety.

BACKGROUND

Healthcare data can be collected in various channels and methods. For example, in some cases, healthcare data is collected by having patients fill in data gathering forms. In some cases, healthcare data is collected by test results. In some cases, healthcare data is collected by video recording, audio recording, and/or images. Organizing different types of healthcare data can be difficult.
Currently, healthcare data is generally organized based on its sources. For example, all test results are stored together. All the patient and healthcare provider communications are stored together. However, the datasets that were generated by different sources are often not organized in any meaningful way. For example, test results for different health issues may be stored together, while the test result of a particular health issue and diagnosis of the particular health issue are not linked together. As another example, datasets of different patients' same health issue or a same patent's repeating health issue are also generally not linked to each other.
Even though a large volume of data is collected, patients or healthcare providers are often drowned in the sea of unorganized data and cannot quickly find relevant information when they need it. For example, when a patient visits a healthcare provider for a repeating health issue, a healthcare provider may need to sift through all the patient's data (some of which is irrelevant to the repeating health issue) to find data related to a previous occurrence of the repeating health issue.
The subject matter claimed herein is not limited to embodiments that solve any disadvantages or that operate only in environments such as those described above. Rather, this background is only provided to illustrate one exemplary technology area where some embodiments described herein may be practiced.

BRIEF SUMMARY

The embodiments described herein are related to a computing system including one or more processors and one or more computer-readable hardware storage devices having stored thereon executable instruction that, when executed by the one or more processors, configure the computing system to perform various acts. The computing system is configured to access a plurality of healthcare datasets, each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact. The artifact includes at least one of (1) textual data, (2) image data, (3) audio data, and/or (4) video data associated with a condition of the patient or care of the patient.
The computing system is further configured to, for each of the plurality of healthcare datasets, label the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels. The computing system is further configured to allow a user to query the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label or a same contextual substitutable label.
In some embodiments, the computing system is further configured to generate a visualization presenting the subset of healthcare datasets or data derived from the subset of healthcare datasets or store the subset of healthcare datasets or data derived from the subset of healthcare datasets in a storage.
In some embodiments, each explicit designation label indicates static information associated with a patient or care of the patient, and each contextual substitutable label indicates substitutable information associated with a condition of the patient or care of the patient. In some embodiments, at least one explicit designation label is linked to one or more contextual substitutable labels.
In some embodiments, the computing system is further configured to identify one or more related healthcare datasets in the plurality of healthcare datasets and store the related healthcare datasets together as a link. In some embodiments, the computing system is further configured to identify one or more related links and store the related links together as a chain. In some embodiments, a chain stores healthcare datasets associated with a same case of a same patient. In some embodiments, querying the plurality of healthcare datasets includes querying one or more datasets in one or more particular links or in one or more particular chains. In some embodiments, the computing system is further configured to anonymize the healthcare datasets before querying the healthcare datasets or presenting a query result.
The embodiments described herein are also related to a computer-implemented method for stitching related healthcare datasets together. The method includes accessing a plurality of healthcare datasets, each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact. The artifact includes at least one of (1) textual data, (2) image data, (3) audio data, or (4) video data associated with a condition of the patient or care of the patient.
The method further includes, for each of the plurality of healthcare datasets, labeling the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels. The method further includes querying the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label or a same contextual substitutable label.
The embodiment described herein are also related to a computer-readable storage device having stored thereon computer-executable instructions. When the computer-executable instructions are executed by a processor of a computing system, the computer-executable instructions configure the computing system to access a plurality of healthcare datasets, each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact. The artifact includes at least one of (1) textual data, (2) image data, (3) audio data, and/or (4) video data associated with a condition of the patient or care of the patient.
The computing system is further configured to, for each of the plurality of healthcare datasets, label the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels. The computing system is further configured to query the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label or a same contextual substitutable label.
Additional features and advantages will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the teachings herein. Features and advantages of the invention may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. Features of the present invention will become more fully apparent from the following description and appended claims or may be learned by the practice of the invention as set forth hereinafter.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to describe the manner in which the above-recited and other advantages and features can be obtained, a more particular description of the subject matter briefly described above will be rendered by reference to specific embodiments, which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments and are not, therefore, to be considered to be limiting in scope, embodiments will be described and explained with additional specificity and details through the use of the accompanying drawings in which:

FIG. 1 illustrates an example data structure of a healthcare dataset (also referred to as a stitch);

FIG. 2 illustrates an example label data structure that is used to label an artifact in a stitch;

FIG. 3 illustrates an example data structure of a link that comprises a plurality of stitches;

FIG. 4 illustrates an example data structure of a chain that comprises a plurality of links;

FIG. 5 illustrates an example embodiment of querying a chain of stitches based on contextual substitutional labels to identify stitches in the chain that share a same contextual substitutional label;

FIG. 6 illustrates an example system in which the principles described herein may be implemented;

FIG. 7 illustrates an example of an embodiment for processing different types of healthcare datasets to identify values of one or more contextual substitutable labels;

FIG. 8 illustrates an example of another embodiment for processing different types of healthcare datasets to identify values of one or more contextual substitutable labels;

FIG. 9 illustrates an example of an embodiment for linking related healthcare datasets together based on a same chief complaint together;

FIG. 10 illustrates an example of an embodiment for identifying related healthcare datasets based on similar chief complaints and/or similarly affected areas;

FIG. 11 illustrates an example of an embodiment for identifying related healthcare datasets based on similar symptoms or lab results; and

FIG. 12 illustrates a flowchart of a method for stitching related healthcare datasets together.

DETAILED DESCRIPTION

Healthcare data can be collected in various channels and methods. For example, in some cases, healthcare data is collected by having patients fill in data gathering forms. In some cases, healthcare data is collected by test results. In some cases, healthcare data is collected by video recording, audio recording, and/or images. Organizing different types of healthcare data can be difficult.
A healthcare provider may include, but is not limited to, any person or entity that provides healthcare or treatment to a patient, is involved in the healthcare or treatment of a patient, and/or is involved with the billing or payment of the healthcare or treatment of a patient. A patient may include, but is not limited to, any person that is a recipient of healthcare or treatment from a healthcare provider or the recipient's family, friends, guardians, caregivers, etc. Examples of healthcare interactions may include, but are not limited to, healthcare provider-healthcare provider interactions (e.g., a discussion between two doctors about a patient's condition or care, a discussion between a dentist and a dental hygienist about a patient's dental plan, a discussion between a medical billing specialist and a medical insurance provider, etc.) and healthcare provider-patient interactions (e.g., a discussion between a physician and a patients about the patient's condition or care, a discussion between a surgeon and the legal guardian of a patient about the patient's upcoming surgery, etc.)
Currently, healthcare data is generally organized based on their sources. For example, all the test results are stored together. All the patient and healthcare provider communications are stored together. However, the datasets that were generated by different sources are often not organized in a meaningful way. Even though a large volume of data is collected, patients or healthcare providers are often drowned in the sea of unorganized data and cannot quickly find relevant information when they need it. For example, when a patient visits a healthcare provider for a repeating health issue, a healthcare provider may need to sift through all the patient's data (some of which is irrelevant to the repeating health issue) to find data related to a previous occurrence of the repeating health issue. As another example, a healthcare provider may want to compile patient data from multiple patients which involves a similar health issue but may need to sift through patient data for each of the multiple patients (some of which may not relate to the similar health issue) to find data related to the similar health issue.
The principles described herein solve the above-described problem by organizing data in a novel data structure. In particular, the principles described herein are related to a computing system, a method, or a computer program product for stitching related healthcare datasets together. A healthcare dataset includes an identifier of a patient, an identifier of a healthcare provider, and an artifact. The artifact can be in different forms, such as (but not limited to) (1) textual data, (2) image data, (3) audio data, and/or (4) video data associated with a condition of the patient. Each healthcare dataset is labeled with one or more explicit designation labels and one or more contextual substitutable labels. These labeled datasets can then be “stitched” together for various purposes. As such, each labeled dataset is also referred to as a “stitch.”
FIG. 1 illustrates an example data structure of a stitch 100. The stitch 100 includes an artifact 110, which may include textual data 112, audio data 114, image data 116, and/or video data 118. The artifact 110 may be generated and/or input by a patient, a healthcare provider, and/or medical equipment, such as an imaging device. For example, when a patient first notices something on their skin, they may decide to take a picture of it. The picture may be stored as an artifact. As another example, a child patient may have a speech disorder, and a parent of the child may take a video of the child talking. The video may also be stored as an artifact. As another example, a patient sees a healthcare provider. The communications between the patient and the healthcare provider may be recorded as an audio file or a video file. The audio file or the video file may also be stored as an artifact.
The artifact 110 is then labeled with a start label 120 (also referred to as a “stitch start label”) and an end label 122 (also referred to as a “stitch end label”). These labels 120, 122 mark the start and the end of the stitch.
The stitch start label 120 and/or the stitch end label 122 also contain additional useful information, including one or more explicit designation labels and one or more contextual substitutable labels. Explicit destination labels indicate static information associated with a patient. The static information contained in an explicit destination label should remain the same for the patient. Such information may include (but is not limited to) family history, birth date, eye color, food allergy, drug allergy, etc.
Contextual substitutable labels indicate contextual and/or substitutable information associated with a condition of the patient. In some embodiments, the contextual substitutable labels indicate data associated with continuing patient encounters based on subjective, objective, assessment, and plan notes (SOAP). Subjective notes may include a patient's chief complaint, a history of the patient's present illness, and pertinent medical history, and/or a current list of the patient's medications. Objective notes may include the patient's vital signs, physical exams, and/or results of any other diagnostics (such as results based on lab tests, imaging, or any other relevant diagnostic information). Assessment notes may include a major diagnosis and clinical stability. Plan notes may include a proposed plan to manage a problem identified based on the assessment notes.
In some embodiments, contextual substitutable labels may be customized by healthcare providers. Each healthcare provider may have their own set of contextual substitutable labels for their patients. In some embodiments, a previous set of contextual substitutable labels may be mutated into a new set of contextual substitutable labels depending on the preference of a healthcare provider. For example, when a patient switches from a first healthcare provider to a second healthcare provider, a first set of contextual substitutable labels of the first healthcare provider may be mutated to a second set of contextual substitutable labels of the second healthcare provider.
In some embodiments, the different artifacts, explicit designation labels and/or contextual substitutable labels are generated at different times, and each artifact and/or label is time-stamped at its creation. The explicit designation labels may be created by a patient; a patient's family, friend, guardian, or caregiver; or a healthcare provider, although the contextual substitutable labels are likely generated by healthcare providers.
FIG. 2 illustrates an example label data structure 200 (which may correspond to the stitch start label 120 or the stitch end label 122 of FIG. 1). The label data structure 200 includes one or more explicit designation labels 212, 214, and one or more contextual substitutable labels 222, 224, 226. The ellipsis 216 represents that there may be any number of explicit designation labels in the label data structure 200. The ellipsis 228 represents that there may be any number of contextual substitutable labels in the label data structure 200.
In some embodiments, some explicit designation labels 212, 214 are linked to one or more contextual substitutable labels 222, 224, 226. For example, explicit designation label 212 is linked to contextual substitutable labels 222, 224; and explicit designation label 214 is linked to contextual substitutable label 226. The labels 212, 214, 222, 224, 226 allow a stitch to be queried by a user to retrieve relevant stitches that the user desires to review. For example, when a healthcare provider or a patient wants to find stitches containing a particular contextual substitutable label, they can query the particular contextual substitutable label to obtain all the stitches containing the particular contextual substitutable label. As such, the stitches obtained from a query can help a healthcare provider or patient to answer health-related questions; to better understand the condition of the patient; and/or improve a plan, process, or treatment of the patient or other patients.
In some embodiments, the multiple related stitches are stored together as a link. FIG. 3 illustrates an example data structure of a link 300. As illustrated in FIG. 3, link 300 includes a plurality of stitches 310, 320, 330, each of which corresponds to the stitch 100 of FIG. 1. The ellipsis 340 represents that there may be any number of stitches in the link 300. The link also includes a link start label 302 and a link end label 304, marking the start and end of the link 300. The link start label 302 and the link end label 304 may also include additional metadata associated with the link.
In some embodiments, multiple related links are also connected together as a chain. FIG. 4 illustrates an example data structure of a chain 400. The chain 400 includes a plurality of links 410, 420, 430, each of which corresponds to the link 300 of FIG. 3. The ellipsis 440 represents that there may be any number of links in the chain 400.
The stitches in a link or a chain can then be queried based on their labels. FIG. 5 illustrates an example embodiment of querying stitches in a chain 510 based on their contextual substitutable labels. As illustrated in FIG. 5, the chain 510 includes two links 520 and 530. Link 520 includes two stitches 522, 524. Link 530 includes three stitches 532, 534, 536. Each stitch includes one or more explicit designation labels (denoted as “E”), and one or more contextual substitutable labels (denoted as “C”).
In particular, stitch 522 includes explicit designation label E1 and contextual substitutable label C1, stitch 524 includes explicit designation label E2 and contextual substitutable label C2, stitch 532 includes explicit designation label E3 and contextual substitutable label E3, and so on and so forth. The stitches 522, 532, and 536 share the same contextual substitutable label C1, as such, stitches 522, 532, and 536 are identified to form a C1 cascade 540. Similarly, stitches 524 and 534 share the same contextual substitutable label C, as such, stitches 524 and 534 are identified together to form a C2 cascade 550.
In some embodiments, many different chains may also be queried based on contextual substitutable labels. In some embodiments, each chain includes stitches associated with a same case of a same patient. In some embodiments, different chains of the same patient may be queried to obtain stitches having a particular contextual substitutable label. In some embodiments, the different chains are associated with patients of a same healthcare provider. In some embodiments, the different chains are associated with patients who have given consent to including their data in the searchable pool. In some embodiments, such a query process removes personally identifiable data from the chains, before presenting the data to a user. The user may be a healthcare provider; a patient; and/or a patient's family, friend, guardian, or caregiver.
For example, cases of a same patient can be queried based on a particular contextual substitutable label to retrieve all the relevant stitches in both cases by either the patient or the healthcare provider. As such, the healthcare provider can visualize and compare the corresponding stitches in two different cases to identify similarities and differences, which can help the healthcare provider to better understand the condition of the patient and/or improve a plan, process, or treatment of the patient or other patients.
In some embodiments, the computer system configured to generate and store stitches, links, and/or chains of data is a server configured to provide a healthcare service application to both patients and healthcare providers. FIG. 6 illustrates an example system 600 in which the principles described herein may be implemented. The system 600 includes a server 610 (which is a networked computing system providing application service or data service to other computing systems), one or more patient devices 620, and one or more healthcare provider devices 630. The patient devices 620 and the healthcare provider devices 630 are configured to communicate with the server 610.
As illustrated in FIG. 6, the server 610 includes or has access to a storage that stores a plurality of healthcare datasets 612. A patient can use a patient device 620 to store and/or access certain healthcare data 612 via a patient agent 622. A healthcare provider can use a healthcare provider device 630 to store and/or access certain healthcare data via a healthcare provider agent 632. The patient device 620 or the healthcare provider device 630 may be a personal computer, a laptop, a mobile phone, a tablet, or any smart device. The patient agent 622 or the healthcare provider agent 632 may be a mobile app or a browser.
Generally, each patient is granted permission to access healthcare data associated with themselves, and each healthcare provider is granted permission to access healthcare data associated with their patients. In embodiments, each patient's healthcare data is organized as stitches (each corresponding to the stitch 100 of FIG. 1), links (each corresponding to the link 300 of FIG. 3), and chains (each corresponding to the chain of FIG. 4). As described above with respect to FIGS. 1-2, each stitch includes a start label and an end label, which contains one or more explicit designation labels and one or more contextual substitutable labels. In some embodiments, a patient is granted permission to enter or edit information associated with the one or more explicit designation labels, and a healthcare provider is granted permission to enter information associated with the one or more contextual substitutional labels. A patient can interact with the patient agent 622 to cause the server 610 to create, modify, and/or review their own healthcare data. Similarly, a healthcare provider can interact with the healthcare provider agent 632 to create, modify, and/or review their patients' healthcare data.
In some embodiments, the patient can interact with the patient agent 622 to cause the server 610 to query their own healthcare data based on their contextual substitutable labels in one or more particular links or chains and cause the subset of stitches to be presented at the patient device 620. A healthcare provider can also interact with the healthcare provider agent 632 to cause the server to query their patients' healthcare data in one or more particular links or chains based on their contextual substitutable labels too and cause the subset of stitches to be presented at the healthcare provider device 630.
In some embodiments, the server 610 also includes one or more artificial intelligence (AI) models 614. In some embodiments, the one or more AI models 614 include language models configured to process certain artifacts in the stitches, such as audio data, image data, and/or audiovisual data to extract textual data. The textual data can then be further processed to obtain meaningful features. These meaningful features can then be further processed to generate or suggest contextual substitutional labels for the corresponding stiches.
In some embodiments, the one or more AI models 614 include object recognition models configured to process audiovisual data and image data. The objects recognized by the AI models 614 can then be further processed to perform diagnosis and/or treatment. The recognized objects or diagnosis can also be used to generate or suggest contextual substitutional labels for the corresponding stitches.
In some embodiments, the one or more AI models 614 may also include a model to identify a patient's sentiment. The identified sentiment may further be used to access the quality of the service provided by the healthcare provider. Such information may also be stored as metadata of a stitch, metadata associated with the patient's profile and/or metadata associated with the healthcare provider's profile.
FIG. 7 illustrates an example embodiment for using AI models 720 (corresponding to AI models 614 of FIG. 6) to process different types of healthcare datasets or artifacts. As illustrated, in some embodiments, a plurality of audiovisual datasets 710 and/or a plurality of audio datasets 760 may be processed by a language model 720 configured to transcribe each of the audio datasets 760 and/or the audiovisual datasets 710 into a textual dataset 730. Further, in some embodiments, the plurality of audiovisual datasets 710 and/or a plurality of image datasets 770 are processed by an object recognizer 724 configured to identify a feature area for at least some of the image datasets and/or the audiovisual datasets.
The textual output of the language model 722 and/or the diagnosis result of the object recognizer 724 can further be processed to generate or suggest one or more contextual substitutable labels. In some embodiments, such machine-generated contextual substitutable labels may be modified or replaced by human users, such as healthcare providers. In some embodiments, the AI models 720 continue to learn from the human interventions to improve their future suggestions. In some embodiments, the AI models 720 are modified based on each healthcare provider, as such, each healthcare provider has their own set of personally customized AI models 720 that are further trained based on their behaviors.
FIG. 8 illustrates an example embodiment for identifying related healthcare datasets 810 to generate links or chains. In some embodiments, these healthcare datasets 810 may be raw datasets or artifacts that have not been processed, such as audiovisual datasets corresponding to a healthcare visit, or an image generated by a test (e.g., CT or MRI). In some embodiments, these healthcare datasets 810 may have been preprocessed via the language model 722 and/or object recognizer 724. In some embodiments, each of the healthcare datasets includes a timestamp 812 indicating a time when the healthcare dataset was created. In some embodiments, the datasets 810 are processed by a machine-learning AI model 820 configured to identify relationships there among. In some embodiments, the healthcare datasets associated with a chief complaint are stored together as a link or a chain. Such a link or chain of datasets may include (but are not limited to) datasets associated with a chief complaint, a prescription, an assessment, a lab result, diagnosis, a procedure performed, a therapy performed, one or more behavior instructions, and/or an affected area 750.
In some embodiments, the computing system is configured to identify one or more related healthcare datasets that share a same attribute 830 and store the one or more related healthcare datasets together. For example, in some embodiments, one or more healthcare datasets that are associated with a same case name or number are stored together.
In some embodiments, a plurality of healthcare datasets associated with a same chief complaint are stored together. FIG. 9 illustrates an example of an embodiment 900, in which a plurality of healthcare datasets associated with a same chief complaint are stored together. The one or more healthcare datasets include a healthcare dataset associated with an initial healthcare visit generating the chief complaint 910, which is linked to a healthcare dataset associated with assessment 920, which is then linked to a healthcare dataset associated with diagnosis 930. Further, in some embodiments, the healthcare dataset associated with diagnosis 930 may further be linked to (1) a healthcare dataset associated with prescription 940, (2) one or more healthcare datasets associated with labs results 950, (3) one or more healthcare datasets associated with one or more procedures performed 960, (4) one or more healthcare datasets associated with one or more therapies performed 970, and/or (5) one or more healthcare datasets associated with one or more behavior instructions 980.
In some embodiments, the plurality of healthcare datasets associated with a same chief complaint are linked together to form a link. A plurality of links that are associated with a case of a patient are then grouped together to form a chain. The patient or the physician can easily review the whole case based on the links and chains of datasets. In some embodiments, each patient may have multiple chains, each of which can further be labeled based on patient's own set of contextual substitutional labels. A patient can query their own healthcare datasets based on their own set of contextual substitutional labels to identify relevant datasets. These relevant datasets can also be presented to the patient as a record, a life story, and/or an album to show family and friends.
Additionally, in some embodiments, the stitches of multiple different cases may further be queried to identify their similarities and relatedness. FIG. 10 illustrates an example of an embodiment 1000 for querying stitches to find related cases. As illustrated in FIG. 10, two healthcare datasets 1010 and 1030 can be identified based on querying a particular contextual substitutional label. For example, in some embodiments, two cases may be identified as related to each other when they share the same or sufficiently similar chief complaint. As another example, in some embodiments, two cases may be identified as related to each other when they share the same affected area.
The multiple cases may be associated with a same patient or different patients, and/or a same healthcare provider or different healthcare providers. When different patients are involved, each patient's healthcare dataset may be anonymized first before querying is performed or the query result is presented. In some embodiments, the patients of the multiple cases involved in the query action are a group of patients that have previously given consent, allowing their healthcare datasets to be included in certain types of query actions. In some embodiments, the stitched data may be presented to each or some of the related patients, and/or each or some of the related healthcare providers.
FIG. 11 further illustrates an example of an embodiment 1100, in which multiple related healthcare datasets are identified based on a query performed based on the related symptom or a lab result. Referring to FIG. 11, healthcare dataset 1112 includes a dataset associated with symptom A, which is associated with (represented by dotted arrow) healthcare dataset 1114, also includes a dataset associated with symptom A. Similarly, the healthcare dataset 1114 is associated with healthcare dataset 1120, which is associated with healthcare dataset 1126, which is associated with 1124, which is associated with 1130, which is associated with healthcare dataset 1132, which is associated with healthcare dataset 1134. Because each of healthcare datasets 1120, 1126, 1124, 1130, 1132, 1134 includes a dataset associated with symptom A, these datasets are identified when a query is performed based on the same symptom A. The symptom A may be indicated as a contextual substitutional label in each of these datasets 1112, 1114, 1120, 1126, 1124, 1130, 1132, 1134.
Further, the healthcare dataset 1112 also includes a dataset associated with lab result B. Thus, the healthcare dataset 1112 is also associated with healthcare dataset 1118, which also includes a dataset associated with lab result B. Similarly, healthcare dataset 1118 is associated with healthcare dataset 1120, which is associated with healthcare dataset 1122, which is associated with healthcare dataset 1128, which is associated with healthcare dataset 1134, which is associated with healthcare dataset 1140, which is associated with dataset 1138. Because each of healthcare datasets 1120, 1122, 1128, 1134, 1140 includes a dataset associated with lab result B, these datasets are identified when a query is performed based on the same lab result B. The lab result B may also be indicated as a contextual substitutional label in each of these datasets 1112, 1118, 1120, 1122, 1128, 1134, 1140, 1138.
Notably, in this example, the healthcare datasets 1116, 1136 are not associated with symptom A or lab result B. Thus, the healthcare dataset 1116, 1136 are not linked to any other healthcare datasets identified based on a query performed based on the same symptom A or a query performed based on the same lab result B.
As illustrated in FIGS. 10 and 11, in some embodiments, a plurality of healthcare datasets associated with a same or similar chief complaint, symptoms, affected area, and/or lab result may be identified based on a query performed based on a same symptom or lab result. However, the principles described herein are not limited to identifying these types of related datasets. For example, in some embodiments, a plurality of healthcare datasets that are associated with artifacts generated via a same type of imaging test and have a high similarity score may also be identified as related to each other. As another example, a plurality of healthcare datasets that are associated with a same affected area and a same patient are identified as related to each other. In some cases, a previous case might have occurred many years ago, but the previous record could help the healthcare provider of the current case to understand the situation and prescribe a targeted treatment.
In some embodiments, the identified related datasets may further be analyzed to generate statistics related to the related datasets. For example, the plurality of healthcare datasets associated with a same or similar chief complaint may further be analyzed to generate statistics associated with the chief complaint; the plurality of healthcare datasets associated with a same or similar symptoms may further be analyzed to generate statistics associated with the symptom; the plurality of healthcare datasets associated with a same or similar diagnosis may further be analyzed to generate statistics associated with the diagnosis; the plurality of healthcare datasets associated with a same affected area may further be analyzed to generate statistics associated with the affected area, and so on and so forth. The statistics may be provided to healthcare providers and/or related patients as reference points.
In some embodiments, when the plurality of the related healthcare datasets are associated with different patients, the computing system is further configured to anonymize the healthcare dataset before conducting the statistical analysis. For example, personally identifiable information may be removed from the datasets. Alternatively, the personally identifiable information may be converted into non-personally identifiable information, such as different age groups, gender groups, professional groups, etc.
The following discussion now refers to a number of methods and method acts that may be performed. Although the method acts may be discussed in a certain order or illustrated in a flow chart as occurring in a particular order, no particular ordering is required unless specifically stated, or required because an act is dependent on another act being completed prior to the act being performed.
FIG. 12 illustrates a flowchart of an example method 1200 for stitching a plurality of related healthcare datasets together. The method 1200 includes accessing a plurality of healthcare datasets (also referred to as a “stitch”), each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact (act 1210). The artifact includes at least one of (1) textual data, (2) image data, (3) audio data, and/or (4) video data associated with a condition of the patient. The method 1200 further includes, for each of the plurality of healthcare datasets, labeling the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels (act 1222).
In some embodiments, each explicit designation label indicates static information associated with a patient or care of the patient, and each contextual substitutable label indicates substitutable information associated with a condition of the patient or care of the patient.
In some embodiments, the method 1200 further includes linking at least one explicit designation label with one or more contextual substitutable labels (act 1224). For example, linking an explicit designation label associated with an allergy with a contextual substitutable label associated with symptoms associated with the allergy.
The method 1200 further includes querying the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label and/or a contextual substitutable label (act 1230). In some embodiments, a visualization is generated to present the subset of healthcare datasets or data derived from the subset of healthcare datasets (act 1240). In some embodiments, the subset of healthcare datasets is stored in a storage that is accessible by a user, such as a healthcare provider or a patient (act 1250). The storage may be a storage of a server (to which a user's mobile device or mobile app has access) or a storage of a mobile device of the user (e.g., a healthcare provider or a patient).
In some embodiments, querying the plurality of healthcare datasets includes querying one or more healthcare datasets that share a particular healthcare provider. In some embodiments, querying the plurality of healthcare datasets includes querying one or more healthcare datasets that share a particular patient.
In some embodiments, the computing system is further configured to identify one or more healthcare datasets as related healthcare datasets and store the related healthcare datasets together as a link. In some embodiments, querying the plurality of healthcare datasets includes querying one or more healthcare datasets in one or more particular links. In some embodiments, the computing system is further configured to identify one or more links as related links and store the related links together as a chain. In some embodiments, querying the plurality of healthcare datasets includes querying one or more healthcare datasets in one or more particular chains.
In some embodiments, a plurality of healthcare datasets associated with a same chief complaint are stored together in a same link or a same chain. For example, the one or more datasets associated with the same chief complaint of a same patient include at least two of (1) a healthcare dataset associated with an initial healthcare visit generating the chief complaint, (2) a healthcare dataset associated with assessment, (3) a healthcare dataset associated with diagnosis, (4) a healthcare dataset associated with prescriptions, (5) a healthcare dataset associated with lab results, (6) a healthcare dataset associated with therapy, or (7) a healthcare dataset associated with behavioral instructions.
In some embodiments, labeling at least some of the plurality of healthcare datasets with contextual substitutable labels is performed via one or more machine-learning models. In some embodiments, the computing system is further configured to anonymize the healthcare datasets before querying the healthcare datasets or presenting the query result to users. In some embodiments, the one or more machine-learning models include a machine-learning language model configured to convert audio or video datasets into textual datasets, or an object recognition model configured to identify an affected area based on a healthcare image dataset or audiovisual dataset.
In some embodiments, the method 1200 further includes substituting one or more contextual substitutable labels with a new set of one or more contextual substitutional labels. The plurality of healthcare datasets can then be queried based on a contextual substitutional label in the new set of one or more contextual substitutional labels. For example, a case of a patient may be transferred from a first healthcare provider to a second healthcare provider. The first healthcare provider may use a first set of contextual substitutional labels to label a dataset, and the second healthcare provider may use a different set of contextual substitutional labels to label the same dataset. In some embodiments, each or some of the contextual substitutional labels in the first sent may be automatically mapped to each or some of the contextual substitutional label in the second set.
In some embodiments, a group of healthcare providers may decide to adopt a different set of contextual substitutional labels due to various reasons. In some embodiments, a patient may define their own set of contextual substitutional labels in user-friendly language, and the healthcare provider's contextual substitutional labels may be automatically mapped to the user-friendly substitutional labels for patients to review.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the described features or acts described above, or the order of the acts described above. Rather, the described features and acts are disclosed as example forms of implementing the claims.
The present invention may comprise or utilize a special-purpose or general-purpose computer system that comprises computer hardware, such as, for example, one or more processors and system memory, as discussed in greater detail below. Configurations within the scope of the present invention also comprise physical and other computer-readable media for carrying or storing computer-executable instructions and/or data structures. Such computer-readable media can be any available media that can be accessed by a general-purpose or special-purpose computer system. Computer-readable media that store computer-executable instructions and/or data structures are computer storage media. Computer-readable media that carry computer-executable instructions and/or data structures are transmission media. Thus, by way of example, and not limitation, configurations of the invention can comprise at least two distinctly different kinds of computer-readable media: computer storage media and transmission media.
Computer storage media are physical storage media that store computer-executable instructions and/or data structures. Physical storage media comprise computer hardware, such as RAM, ROM, EEPROM, solid state drives (“SSDs”), flash memory, phase-change memory (“PCM”), optical disk storage, magnetic disk storage or other magnetic storage devices, or any other hardware storage device(s) which can be used to store program code in the form of computer-executable instructions or data structures, which can be accessed and executed by a general-purpose or special-purpose computer system to implement the disclosed functionality of the invention.
Transmission media can comprise a network and/or data links which can be used to carry program code in the form of computer-executable instructions or data structures, and which can be accessed by a general-purpose or special-purpose computer system. A “network” is defined as one or more data links that enable the transport of electronic data between computer systems and/or modules and/or other electronic devices. When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or a combination of hardwired or wireless) to a computer system, the computer system may view the connection as transmission media. Combinations of the above should also be comprised within the scope of computer-readable media.
Further, upon reaching various computer system components, program code in the form of computer-executable instructions or data structures can be transferred automatically from transmission media to computer storage media (or vice versa). For example, computer-executable instructions or data structures received over a network or data link can be buffered in RAM within a network interface module (e.g., a “NIC”), and then eventually transferred to computer system RAM and/or to less volatile computer storage media at a computer system. Thus, it should be understood that computer storage media can be comprised in computer system components that also (or even primarily) utilize transmission media.
Computer-executable instructions comprise, for example, instructions and data which, when executed at one or more processors, cause a general-purpose computer system, special-purpose computer system, or special-purpose processing device to perform a certain function or group of functions. Computer-executable instructions may be, for example, binaries, intermediate format instructions such as assembly language, or even source code.
Those skilled in the art will appreciate that the invention may be practiced in network computing environments with many types of computer system configurations, including, personal computers, desktop computers, laptop computers, message processors, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, mobile telephones, PDAs, tablets, pagers, routers, switches, and the like. The invention may also be practiced in distributed system environments where local and remote computer systems, which are linked (either by hardwired data links, wireless data links, or by a combination of hardwired and wireless data links) through a network, both perform tasks. As such, in a distributed system environment, a computer system may comprise a plurality of constituent computer systems. In a distributed system environment, program modules may be located in both local and remote memory storage devices.
Those skilled in the art will also appreciate that the invention may be practiced in a cloud-computing environment. Cloud computing environments may be distributed, although this is not required. When distributed, cloud computing environments may be distributed internationally within an organization and/or have components possessed across multiple organizations. In this description and the following claims, “cloud computing” is defined as a model for enabling on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services). The definition of “cloud computing” is not limited to any of the other numerous advantages that can be obtained from such a model when properly deployed.
A cloud-computing model can be composed of various characteristics, such as on-demand self-service, broad network access, resource pooling, rapid elasticity, measured service, and so forth. A cloud-computing model may also come in the form of various service models such as, for example, Software as a Service (“SaaS”), Platform as a Service (“PaaS”), and Infrastructure as a Service (“IaaS”). The cloud-computing model may also be deployed using different deployment models such as private cloud, community cloud, public cloud, hybrid cloud, and so forth.
Some configurations, such as a cloud-computing environment, may comprise a system that comprises one or more hosts that are each capable of running one or more virtual machines. During operation, virtual machines emulate an operational computing system, supporting an operating system and perhaps one or more other applications as well. In some configurations, each host comprises a hypervisor that emulates virtual resources for the virtual machines using physical resources that are abstracted from view of the virtual machines. The hypervisor also provides proper isolation between the virtual machines. Thus, from the perspective of any given virtual machine, the hypervisor provides the illusion that the virtual machine is interfacing with a physical resource, even though the virtual machine only interfaces with the appearance (e.g., a virtual resource) of a physical resource. Examples of physical resources including processing capacity, memory, disk space, network bandwidth, media drives, and so forth.
The present invention may be embodied in other specific forms without departing from its spirit or characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

What is claimed is:

1. A computing system comprising:

one or more processors; and

one or more computer-readable hardware storage devices having stored thereon computer-executable instructions that are structured such that, when executed by the one or more processors, cause the computing system to perform at least:

access a plurality of healthcare datasets, each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact, the artifact including at least one of (1) textual data, (2) image data, (3) audio data, or (4) video data associated with a condition of the patient or care of the patient;

for each of the plurality of healthcare datasets, label the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels; and

query the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label or a contextual substitutable label.

2. The computing system of claim 1, the computing system further configured to:

generate a visualization presenting the subset of healthcare datasets or data derived from the subset of healthcare datasets; or

store the subset of healthcare datasets that share a same explicit designation label or a contextual substitutable label in a storage.

3. The computing system of claim 1, wherein each explicit designation label indicates static information associated with a patient or care of the patient, and each contextual substitutable label indicates substitutable information associated with a condition of the patient or care of the patient.

4. The computing system of claim 1, wherein the computing system is further configured to link at least one of the one or more explicit designation labels to one or more contextual substitutable labels.

5. The computing system of claim 1, wherein querying the plurality of healthcare datasets includes querying one or more healthcare datasets that share a same particular healthcare provider or a particular patient.

6. The computing system of claim 5, wherein the computing system is further configured to:

identify one or more healthcare datasets as related healthcare datasets; and

store the related healthcare datasets together as a link.

7. The computing system of claim 6, wherein querying the plurality of healthcare datasets includes querying one or more healthcare datasets in one or more particular links.

8. The computing system of claim 6, wherein the computing system is further configured to:

identify one or more links as related links; and

store the related links together as a chain.

9. The computing system of claim 8, wherein querying the plurality of healthcare datasets includes querying one or more healthcare datasets in one or more particular chains.

10. The computing system of claim 8, wherein a plurality of healthcare datasets associated with a particular chief complaint are stored together in a same link or a same chain, the one or more healthcare datasets associated with the particular chief complaint comprising at least two of: (1) a healthcare dataset associated with an initial healthcare visit generating the particular chief complaint, (2) a healthcare dataset associated with assessment, (3) a healthcare dataset associated with diagnosis, (4) a healthcare dataset associated with prescriptions, (5) a healthcare dataset associated with lab results, (6) a healthcare dataset associated with therapy, or (7) a healthcare dataset associated with behavioral instructions.

11. The computing system of claim 1, wherein labeling at least some of the plurality of healthcare datasets with contextual substitutable labels is performed via one or more machine-learning models.

12. The computing system of claim 11, wherein the one or more machine-learning models include a machine-learning language model configured to convert audio or video datasets into textual datasets, or an object recognition model configured to identify an affected area based on a healthcare image dataset or a healthcare audiovisual dataset.

13. The computing system of claim 1, wherein the computing system is further configured to substitute one or more contextual substitutable labels with a new set of one or more contextual substitutional labels.

14. The computing system of claim 13, wherein the computing system is further configured to query the plurality of healthcare datasets based on a contextual substitutional label in the new set of one or more contextual substitutional labels.

15. The computing system of claim 1, wherein the computing system further configured to anonymize the healthcare datasets before querying the healthcare datasets.

16. A method implemented at a computing system for stitching a plurality of related healthcare datasets, the method comprising:

accessing a plurality of healthcare datasets, each of which includes an identifier of a patient, an identifier of a healthcare provider, and an artifact, the artifact including at least one of (1) textual data, (2) image data, (3) audio data, or (4) video data associated with a condition of the patient or care of the patient;

for each of the plurality of healthcare datasets, labeling the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels, each explicit designation label indicating static information associated with a patient, and each contextual substitutable label indicating substitutable information associated with a condition of the patient; and

querying the plurality of healthcare datasets to identify a subset of healthcare datasets that share a same explicit designation label or a contextual substitutable label.

17. The method of claim 16, the method further comprising:

generating a visualization presenting the subset of healthcare datasets or data derived from the subset of healthcare datasets; or

storing the subset of healthcare datasets that share a same explicit designation label or a contextual substitutable label in a storage.

18. The method of claim 16, wherein each explicit designation label indicates static information associated with a patient or care of the patient, and each contextual substitutable label indicates substitutable information associated with a condition of the patient or care of the patient.

19. The method of claim 15, wherein the computing system is further configured to link at least one of the one or more explicit designation labels to one or more contextual substitutable labels.

20. A computer program product comprising one or more hardware storage devices having stored thereon computer-executable instructions that are structured such that, when the computer-executable instructions are executed by one or more processors of a computing system, the computer-executable instructions configure the computing system to perform at least:

for each of the plurality of healthcare datasets,

label the healthcare dataset with one or more explicit designation labels and one or more contextual substitutable labels, each explicit designation label indicating static information associated with a patient or care of the patient, and each contextual substitutable label indicating substitutable information associated with a condition of the patient or care of the patient;

store related one or more healthcare datasets in a link;

store related one or more links in a chain; and

query the plurality of healthcare datasets in one or more particular links or one or more particular chains to identify a subset of healthcare datasets that share a same contextual substitutable label or a same explicit designation label.