CPC A61B 5/165 (2013.01) [G06F 18/295 (2023.01); G16H 20/70 (2018.01)] | 16 Claims |
1. A computer implemented method for assessing mental health of a person, the method comprising:
recording, by a processor of a user's mobile device, digital activity data of the user on the mobile device for a predetermined duration, wherein the digital activity data does not include personal content of the user;
applying, by the processor of the user's mobile device, the digital activity data to a plurality of machine learning models, wherein each of the machine learning models represents a digital phenotype for a different mental health condition;
determining one or more behavioral profiles to train the machine learning models, wherein the behavioral profiles are clustered based on digital phenotypes indicative of at least one mental health condition;
determining, by the processor of the user's mobile device, a similarity score for each mental health condition corresponding to each of the plurality of machine learning models based on comparing an output of the respective machine learning model to a digital phenotype for the corresponding mental health condition, wherein the similarity score is a confidence level for a match to someone with that specific mental health condition;
outputting, at a user's display, the at least one mental health similarity score; and
predicting, by the processor, at least one mental health condition of the user based on the calculated similarity scores,
wherein the predetermined duration is modified in real-time depending upon the at least one mental health condition.
|