US 11,942,194 B2
Systems and methods for mental health assessment
Elizabeth E. Shriberg, Berkeley, CA (US); Michael Aratow, Mountain View, CA (US); Mainul Islam, San Francisco, CA (US); Amir Hossein Harati Nejad Torbati, Toronto (CA); Tomasz Rutowski, Gdansk (PL); David Lin, Foster City, CA (US); Yang Lu, Waterloo (CA); Farshid Haque, San Francisco, CA (US); and Robert D. Rogers, Pleasanton, CA (US)
Assigned to Ellipsis Health, Inc., San Francisco, CA (US)
Filed by Ellipsis Health, Inc., San Francisco, CA (US)
Filed on Feb. 14, 2022, as Appl. No. 17/671,337.
Application 17/671,337 is a continuation of application No. 17/400,066, filed on Aug. 11, 2021.
Application 17/400,066 is a continuation of application No. 17/129,859, filed on Dec. 21, 2020, granted, now 11,120,895, issued on Sep. 14, 2021.
Application 17/129,859 is a continuation of application No. 16/918,624, filed on Jul. 1, 2020, abandoned.
Application 16/918,624 is a continuation of application No. 16/560,720, filed on Sep. 4, 2019, granted, now 10,748,644, issued on Aug. 18, 2020.
Application 16/560,720 is a continuation of application No. 16/523,298, filed on Jul. 26, 2019, abandoned.
Application 16/523,298 is a continuation of application No. PCT/US2019/037953, filed on Jun. 19, 2019.
Claims priority of provisional application 62/755,361, filed on Nov. 2, 2018.
Claims priority of provisional application 62/755,356, filed on Nov. 2, 2018.
Claims priority of provisional application 62/754,534, filed on Nov. 1, 2018.
Claims priority of provisional application 62/754,547, filed on Nov. 1, 2018.
Claims priority of provisional application 62/754,541, filed on Nov. 1, 2018.
Claims priority of provisional application 62/749,672, filed on Oct. 24, 2018.
Claims priority of provisional application 62/749,669, filed on Oct. 23, 2018.
Claims priority of provisional application 62/749,654, filed on Oct. 23, 2018.
Claims priority of provisional application 62/749,663, filed on Oct. 23, 2018.
Claims priority of provisional application 62/749,113, filed on Oct. 22, 2018.
Claims priority of provisional application 62/733,552, filed on Sep. 19, 2018.
Claims priority of provisional application 62/733,568, filed on Sep. 19, 2018.
Claims priority of provisional application 62/687,176, filed on Jun. 19, 2018.
Prior Publication US 2022/0165371 A1, May 26, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G16H 10/20 (2018.01); A61B 5/00 (2006.01); A61B 5/16 (2006.01); G09B 19/00 (2006.01); G10L 25/66 (2013.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01); G06F 16/24 (2019.01); G10L 15/18 (2013.01)
CPC G16H 10/20 (2018.01) [A61B 5/164 (2013.01); A61B 5/165 (2013.01); A61B 5/4803 (2013.01); G09B 19/00 (2013.01); G10L 25/66 (2013.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01); A61B 5/4088 (2013.01); A61B 5/7275 (2013.01); G06F 16/24 (2019.01); G10L 15/18 (2013.01)] 27 Claims
OG exemplary drawing
 
1. A method of assessing a mental health state or predicting a risk level of a subject having one or more mental health conditions, the method comprising:
(a) eliciting at least one conversational-style spoken response from the subject, at least in part by engaging the subject in a spoken conversation at least in part by presenting one or more open-ended questions to the subject, wherein the at least one conversational-style spoken response comprises voice audio data;
(b) applying one or more deep learning systems to the at least one conversational-style spoken response to extract feature representations from the at least one conversational-style spoken response, wherein the feature representations comprise one or more acoustic features and/or language features;
(c) using one or more machine learning models to assess the mental health state or predict the risk level of the subject based at least on the feature representations and generate a prediction confidence of the mental health state or the risk level of the subject, wherein the one or more machine learning models comprise at least one of an acoustic model or a natural language processing (NLP) model;
(d) generating at least one metric in real-time by weighting or selecting the mental health state or the risk level of the subject assessed in (c) by each of the one of more machine learning models based on the prediction confidence, wherein the at least one metric comprises one or more scaled scores that are indicative of severity of the one or more mental health conditions, wherein the one or more scaled scores are generated using at least in part assessments by at least one of the one or more machine learning models; and
(e) providing the at least one metric associated with the subject to one or more clinical call centers, telehealth platforms and/or remote patient monitoring applications, for integration and establishing of one or more care pathways for the subject.