# US 2018/0025126 A1 · System and method for predictive modeling and adjustment of behavioral health

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2018/0025126 A1
- Document ID: US20180025126A1
- Status: Published application
- Kind: A1
- Application number: 15/647,399
- Filed: 2017-07-12
- Published: 2018-01-25
- Earliest claimed: 2016-07-20
- Assignees: Synchronous Health, Inc.
- Applicants: Synchronous Health, Inc.
- Inventors: Guy Barnard; Lisa D. Henderson; Katherine A. Lohr
- Web page: https://patentgrep.com/patent/US20180025126A1

## CPC

- A61B 5/0022: {Monitoring a patient using a global network, e.g. telephone networks, internet}
- A61B 5/1113: {Local tracking of patients, e.g. in a hospital or private home}
- A61B 5/1118: {Determining activity level}
- A61B 5/165: {Evaluating the state of mind, e.g. depression, anxiety}
- A61B 5/4833: {Assessment of subject's compliance to treatment}
- A61B 5/7465: {Arrangements for interactive communication between patient and care services, e.g. by using a telephone network (telemetry of measured physiological signal A61B5/0002)}
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/50: for simulation or modelling of medical disorders
- G16H 15/00: ICT specially adapted for medical reports, e.g. generation or transmission thereof
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

A cloud-based system and method for predictive modeling and positive adjustment of behavioral health are disclosed. The system includes sensors collecting data associated with subject location and activity, and linked to a subject computing device. The system translates data aggregated from the data sources into state information, and iteratively updates, via the translated state information, a de-identified contextual model for the subject which in an embodiment may be a Life Context Graph as described herein. An end point server compares the de-identified contextual model to a de-identified aggregate of peer-based contextual models, wherein data security and privacy is preserved, and the system further iteratively updates the subject contextual model based thereon. The system accordingly identifies behavioral trigger actions based on the collected data and/or the updated subject contextual model, and generates a predetermined clinical response corresponding to the identified behavioral trigger action.

## Claims

### Claim 1

A system for predictive modeling and positive adjustment of behavioral health, the system comprising:
one or more data sources comprising one or more sensors configured to collect data associated with subject location and activity;
a subject computing device associated with at least the one or more sensors and linked to a communications network, and further comprising a processor configured to:
translate data aggregated from the one or more data sources into state information,
iteratively update, via the translated state information, a de-identified contextual model associated with the subject,
transmit the de-identified contextual model to an end point server configured to compare the subject contextual model to a de-identified aggregate of one or more peer-based contextual models and further iteratively update the subject contextual model based on said comparison,
identify a behavioral trigger action by the subject based on one or more of the collected data associated with subject activity and the updated subject contextual model, and
generate a predetermined clinical response corresponding to the identified behavioral trigger action.

### Claim 2 (depends on 1)

The system of claim 1, further comprising a clinician decision support interface configured for display and user interaction via a clinician computing device,
wherein the clinician decision support interface comprises software tools enabling selective generation of a trigger comprising each of an IF or WHEN component, a state component and an action component,
wherein the processor associated with the subject computing device is configured to identify the behavioral trigger action by:
identifying the presence of the state component via one or more of the collected data associated with subject activity and the updated subject contextual model, and
causing implementation of the action component associated with the trigger.

### Claim 3 (depends on 2)

The system of claim 2, wherein the state component comprises a combination of one or more states developed from the translated data aggregated from the one or more data sources into state information.

### Claim 4 (depends on 2)

The system of claim 2, wherein the action component comprises one or more of an interactive messaging application executed via a graphical user interface associated with the subject computing device and an automated regulation of settings for one or more home environment devices associated with the subject.

### Claim 5 (depends on 1)

The system of claim 1, wherein the processor associated with the subject computing device is further configured to employ
homomorphic encryption tools to de-identify the subject contextual model at one or more of the subject computing device and the end point server, and
block chain signature tools to manage iterations or states associated with the contextual model.

### Claim 6 (depends on 1)

The system of claim 1, wherein the data sources comprise one or more remote state tracking devices linked to a hosted server and configured to provide one or more remote origins of state information, wherein the hosted server is configured to track a last known state of the state information when the remote state tracking device is not currently accessible via the communications network.

### Claim 7 (depends on 1)

The system of claim 1, further comprising a state feedback loop wherein the processor associated with the subject computing device is further configured to generate iterative updates to the contextual model based on subject action or inaction corresponding to the clinical behavioral response.

### Claim 8 (depends on 7)

The system of claim 7, wherein the processor associated with the subject computing device is further configured to generate, responsive to subject inaction associated with the state feedback loop, one or more of a predetermined secondary clinical behavioral response and an interactive query string via a user interface for the subject computing device.

### Claim 9

A method for predictive modeling and positive adjustment of behavioral health, the method comprising:
continuously collecting and aggregating, via at least one or more sensors, data associated with subject location and activity;
translating the aggregated data into state information;
iteratively updating, via the translated state information, a de-identified contextual model associated with the subject;
transmitting the de-identified contextual model to a centralized location via a communications network;
comparing the subject contextual model to a de-identified aggregate of one or more peer-based contextual models;
iteratively updating the subject contextual model based on said comparison;
identifying a behavioral trigger action by the subject based on one or more of the collected data associated with subject activity and the updated subject contextual model; and
generating a predetermined clinical response corresponding to the identified behavioral trigger action.

### Claim 10 (depends on 9)

The method of claim 9, further comprising a clinician decision support interface configured for display and user interaction via a clinician computing device,
enabling, via a clinician decision support interface, selective generation of a trigger comprising each of an IF or WHEN component, a state component and an action component; and
identifying the behavioral trigger action by:
identifying the presence of the state component via one or more of the collected data associated with subject activity and the updated subject contextual model, and
causing implementation of the action component associated with the trigger.

### Claim 11 (depends on 10)

The method of claim 10, wherein the state component comprises a combination of one or more states developed from the translated data into state information.

### Claim 12 (depends on 10)

The method of claim 10, wherein the action component comprises an interactive messaging application executed via a graphical user interface associated with a subject computing device.

### Claim 13 (depends on 10)

The method of claim 10, wherein the action component comprises an automated regulation of settings for one or more home environment devices associated with the subject.

### Claim 14 (depends on 10)

The method of claim 10, further comprising:
employing homomorphic encryption tools to de-identify the subject contextual model at one or more of the subject computing device and the end point server, and
employing block chain signature tools to manage iterations or states associated with the contextual model.

### Claim 15 (depends on 14)

The method of claim 14, further comprising generating iterative updates to the contextual model based on subject action or inaction corresponding to the clinical behavioral response.

### Claim 16 (depends on 14)

The method of claim 14, further comprising generating, responsive to subject inaction, one or more of a predetermined secondary clinical behavioral response and an interactive query string via a user interface for the subject computing device.

### Claim 17

A system for predictive modeling and positive adjustment of behavioral health, the system comprising:
means for aggregating data about a subject from a plurality of connected devices;
means for continuously translating the aggregated data into an ongoing series of states defining a Life Context Graph;
means for comparing the Life Context Graph to a de-identified aggregate of one or more peer-based Life Context Graphs;
means for predicting potential behavioral health risks for the subject; and
means for generating automated queries and messages to the subject with respect to behavioral health improvements and adjustments corresponding to the predicted potential risks.

### Claim 18 (depends on 17)

The system of claim 17, wherein the means for comparing the Life Context Graph to a de-identified aggregate of one or more peer-based Life Context Graphs comprises employing differential privacy and homomorphic encryption models to transmit the subject's aggregated data and states to an end-point aggregator.

### Claim 19 (depends on 18)

The system of claim 18, further comprising means for ensuring authenticity of data in untrusted networks across graph iterations or states.

### Claim 20 (depends on 17)

The system of claim 17, wherein the means for generating automated queries and messages further comprises asynchronous program execution for generation of a linear workflow with respect to trigger designation and specified actions.