# US 11,908,585 B2 · Method for modeling behavior and depression state

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

- Number: US 11,908,585 B2
- Document ID: US11908585B2
- Status: Granted
- Kind: B2
- Application number: 17/517,561
- Filed: 2021-11-02
- Published: 2022-02-24
- Granted: 2024-02-20
- Earliest claimed: 2012-08-16
- Assignees: OrangeDot, Inc.
- Applicants: OrangeDot, Inc.
- Inventors: Sai Moturu; Anmol Madan; Shishir Dash; Karim Wahba; Gourab De
- Web page: https://patentgrep.com/patent/US11908585B2

## CPC

- G16H 50/50: for simulation or modelling of medical disorders
- G16H 10/20: for electronic clinical trials or questionnaires
- G16H 40/63: for local operation
- G16H 40/67: for remote operation
- G16Z 99/00: Subject matter not provided for in other main groups of this subclass
- G16H 20/10: relating to drugs or medications, e.g. for ensuring correct administration to patients
- G16H 20/70: relating to mental therapies, e.g. psychological therapy or autogenous training
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16Z: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR

## Abstract

A method and system for modeling behavior and depression state of an individual, the method comprising: receiving a log of use dataset associated with communication behavior of the individual during a time period; receiving a supplementary dataset characterizing activity of the individual during the time period; receiving a survey dataset including responses, to at least one of a set of depression-assessment surveys, associated with a set of time points of the time period; generating a predictive analysis of a depression-risk state of the individual associated with at least a portion of the time period, from at least one of the log of use dataset, the supplementary dataset, and the survey dataset; and generating an alert upon detection that a set of parameters from the predictive analysis of the depression-risk state satisfy a threshold condition.

## Claims

### Claim 1

A method for improving depression state determination and monitoring for a user, the method comprising:
receiving a first communication dataset from an application executing on a mobile communication device of the user, wherein the first communication dataset comprises a first set of message transcripts associated with a first time period;
with a first model, deriving a first textual content associated with the first set of message transcripts;
with a second model, wherein the second model comprises a trained predictive model, determining a first set of depression risk parameters associated with the user based on the first textual content;
assessing the first set of depression risk parameters to determine a first depression risk state associated with the user;
in response to and based on the first depression risk state, determining and initiating provision of a first therapeutic intervention at the mobile communication device, the first therapeutic intervention operable to improve the depression state of the user;
receiving a second communication dataset from the application, wherein the second communication dataset comprises a second set of message transcripts associated with a second time period, the second time period occurring later in time than the first time period;
with the first model, deriving a second textual content associated with the second set of message transcripts;
with the second model, determining a second depression risk state associated with the user based on the second textual content;
in response to the second depression risk state and based on a change from the first depression risk state to the second depression risk state, determining and initiating provision of a second therapeutic intervention at the mobile communication device, the second therapeutic intervention operable to improve the depression state of the user;
updating the second model based on the change from the first depression risk state to the second depression risk state, wherein updating the second model is operable to increase a predictive accuracy of the second model for the user; and
with the updated second model, determining a third depression risk state associated with the user based on a third textual content.

### Claim 2 (depends on 1)

The method of claim 1, wherein the first model is a trained model.

### Claim 3 (depends on 1)

The method of claim 1, wherein the textual content comprises a set of content features, the set of content features indicative of at least one of a set of negative sentiments and a set of positive sentiments associated with the set of message transcripts.

### Claim 4 (depends on 1)

The method of claim 1, further comprising, with the first model, determining a set of temporal parameters associated with the first communication dataset, wherein the first set of depression risk parameters is further determined based on the set of temporal parameters.

### Claim 5 (depends on 4)

The method of claim 4, wherein the set of temporal parameters comprises at least one of: a frequency associated with receipt of message transcripts of the first set of message transcripts; and a time associated with receipt of message transcripts of the first set of message transcripts.

### Claim 6 (depends on 1)

The method of claim 1, further comprising, with the first model, determining a set of message lengths associated with message transcripts of the first set of messages, wherein the set of depression risk parameters is further determined based on the set of message lengths.

### Claim 7 (depends on 1)

The method of claim 1, wherein assessing the first set of depression risk parameters to determine the first depression risk state comprises comparing the first set of depression risk parameters with a set of thresholds.

### Claim 8

A method for improving depression state determination and monitoring for a user, the method comprising:
receiving a first communication dataset from an application executing on a mobile communication device of the user;
with a trained predictive model, determining a first set of depression risk parameters associated with the user based on the first communication dataset;
assessing the first set of depression risk parameters to determine a first depression risk state associated with the user;
in response to and based on the first depression risk state, determining and initiating provision of a first therapeutic intervention at the mobile communication device, the first therapeutic intervention operable to improve the depression state of the user;
receiving a second communication dataset from the application, wherein the second communication dataset is associated with a second time period, the second time period occurring later in time than the first time period;
with the trained predictive model, determining a second depression risk state associated with the user based on the second communication dataset;
in response to and based on the second depression risk state, determining and initiating provision of a second therapeutic intervention at the mobile communication device, the second therapeutic intervention operable to improve the depression state of the user; and
updating the trained predictive model based on the second depression risk state wherein updating the trained predictive model is operable to increase a predictive accuracy of the trained predictive model to determine a depression risk state associated with the user; and
with the updated trained predictive model, determining a third depression risk state associated with the user based on a third communication dataset.

### Claim 9 (depends on 8)

The method of claim 8, wherein the first communication dataset comprises a first set of message transcripts and wherein the second communication dataset comprises a second set of message transcripts.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
determining a first textual content associated with the first set of message transcripts, wherein the first set of depression risk parameters is determined based on the first textual content; and
determining a second textual content associated with the second set of message transcripts, wherein the second set of depression risk parameters is determined based on the second textual content.

### Claim 11 (depends on 10)

The method of claim 10, wherein each of the first and second textual contents is associated with at least one of a positive and negative sentiment.

### Claim 12 (depends on 8)

The method of claim 8, further comprising determining a change from the first depression risk state to the second depression risk state, wherein the trained predictive model is updated based on the change.

### Claim 13 (depends on 12)

The method of claim 12, further comprising receiving a survey dataset from the user during the second time period, wherein the second depression risk state is further determined based on the survey dataset.

### Claim 14 (depends on 8)

The method of claim 8, further comprising determining a set of features associated with the user, wherein the trained predictive model is determined based on the set of features.

### Claim 15 (depends on 14)

The method of claim 14, wherein the trained predictive model is further determined based on aggregated data from a set of multiple users, wherein the set of multiple users share one or more of the set of features with the user.

### Claim 16 (depends on 8)

The method of claim 8, further comprising determining a set of temporal parameters associated with the first communication dataset, wherein the first set of depression risk parameters is further determined based on the set of temporal parameters.

### Claim 17 (depends on 16)

The method of claim 16, wherein the set of temporal parameters comprises at least one of: a frequency associated with receipt of a set of messages of the first communication dataset; and a time associated with receipt of the set of messages.

### Claim 18 (depends on 8)

The method of claim 8, wherein assessing the first set of depression risk parameters to determine the first depression risk state comprises comparing the first set of depression risk parameters with a set of thresholds.