Systems and methods for mental health assessment

Abstract

The present disclosure provides systems and methods for assessing a mental state of a subject in a single session or over multiple different sessions, using for example an automated module to present and/or formulate at least one query based in part on one or more target mental states to be assessed. The query may be configured to elicit at least one response from the subject. The query may be transmitted in an audio, visual, and/or textual format to the subject to elicit the response. Data comprising the response from the subject can be received. The data can be processed using one or more individual, joint, or fused models. One or more assessments of the mental state associated with the subject can be generated for the single session, for each of the multiple different sessions, or upon completion of one or more sessions of the multiple different sessions.

1 .- 20 . (canceled) 
     
     
         21 . A method, comprising:
 (a) obtaining first speech data from a subject at a first time point;   (b) using an acoustic model and a natural language processing (NLP) model, processing said first speech data from said subject to generate a first metric that is indicative of whether said subject has said behavioral or mental condition at said first time point;   (c) obtaining second speech data from said subject at a second time point, wherein said second time point is after said first time point;   (d) using said acoustic model and said NLP model, processing said second speech data from said subject to generate a second metric that is indicative of whether said subject has said behavioral or mental condition at said second time point;   (e) displaying, in a graphical user interface of an electronic device, a comparison of said first metric and said second metric to enable said subject or another user to track a progression of said behavioral or mental health condition over time.   
     
     
         22 . The method of  claim 21 , wherein said comparison comprises a plot comprising said first metric and said second metric. 
     
     
         23 . The method of  claim 21 , further comprising displaying on said graphical user interface a qualitative assessment associated with said first metric or said second metric. 
     
     
         24 . The method of  claim 21 , further comprising displaying one or more topics identified in said first speech data or said second speech data. 
     
     
         25 . The method of  claim 21 , further comprising displaying a personalized health recommendation on said graphical user interface of said electronic device. 
     
     
         26 . The method of  claim 25 , wherein said personalized health recommendation is a referral to a healthcare provider. 
     
     
         27 . The method of  claim 21 , further comprising transmitting an alert to a healthcare provider in response to said second metric satisfying a condition. 
     
     
         28 . The method of  claim 21 , further comprising establishing a baseline profile for said subject. 
     
     
         29 . The method of  claim 28 , further comprising calibrating said first metric or said second metric based on said baseline profile. 
     
     
         30 . The method of  claim 21 , wherein said acoustic model or said NLP model is personalized for a demographic group to which said subject belongs. 
     
     
         31 . The method of  claim 21 , wherein (a) or (c) comprises prompting said subject to an answer a question related to a mental state or mood of said subject. 
     
     
         32 . The method of  claim 21 , wherein (a) or (c) is performed before, during, or after a clinical encounter with a healthcare provider. 
     
     
         33 . The method of  claim 21 , wherein said acoustic model and said NLP model are deep neural networks. 
     
     
         34 . The method of  claim 33 , wherein said deep neural networks are trained on a plurality of speech samples generated by a plurality of other subjects. 
     
     
         35 . The method of  claim 34 , wherein each of said plurality of speech samples comprises a label that indicates that the other subject that generated said speech sample (i) has, to some level, said behavioral or mental condition or (ii) does not have said behavioral or mental condition. 
     
     
         36 . The method of  claim 35 , wherein said label is based on a clinical diagnosis. 
     
     
         37 . The method of  claim 35 , wherein said label is based on a clinically validated survey or questionnaire. 
     
     
         38 . The method of  claim 21 , wherein said first metric and said second metric are scaled scores. 
     
     
         39 . The method of  claim 21 , wherein said another user is a healthcare provider. 
     
     
         40 . The method of  claim 39 , wherein said healthcare provider is a psychologist, a psychiatrist, or a therapist. 
     
     
         41 . The method of  claim 21 , further comprising connecting said subject to a healthcare provider through said graphical user interface of said electronic device in response to said first metric or said second metric satisfying a condition.