US 12,263,011 B2
Treatment of depression using machine learning
Amit Etkin, Palo Alto, CA (US); Madhukar Trivedi, Dallas, TX (US); and Wei Wu, Palo Alto, CA (US)
Assigned to Department of Veternas Affairs, Washington, DC (US); and Board of Regents, The University of Texas System, Austin, TX (US)
Appl. No. 17/282,624
Filed by The Board of Trustees of the Leland Stanford Junior University, Stanford, CA (US); and Board of Regents, The University of Texas System, Austin, TX (US)
PCT Filed Oct. 15, 2019, PCT No. PCT/US2019/056395
§ 371(c)(1), (2) Date Apr. 2, 2021,
PCT Pub. No. WO2020/081609, PCT Pub. Date Apr. 23, 2020.
Claims priority of provisional application 62/838,080, filed on Apr. 24, 2019.
Claims priority of provisional application 62/745,890, filed on Oct. 15, 2018.
Prior Publication US 2021/0353224 A1, Nov. 18, 2021
Int. Cl. G01N 33/48 (2006.01); A61B 5/00 (2006.01); A61B 5/16 (2006.01); A61B 5/245 (2021.01); A61B 5/374 (2021.01); A61B 5/377 (2021.01); G01N 33/50 (2006.01); G01R 33/48 (2006.01); G06N 20/00 (2019.01); G16H 30/20 (2018.01); G16H 30/40 (2018.01); G16H 40/67 (2018.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01); A61N 1/36 (2006.01); A61N 1/38 (2006.01); A61N 2/00 (2006.01); G16H 20/10 (2018.01); G16H 20/30 (2018.01); G16H 20/40 (2018.01); G16H 20/70 (2018.01)
CPC A61B 5/4848 (2013.01) [A61B 5/0075 (2013.01); A61B 5/165 (2013.01); A61B 5/168 (2013.01); A61B 5/245 (2021.01); A61B 5/374 (2021.01); A61B 5/377 (2021.01); A61B 5/4082 (2013.01); A61B 5/4088 (2013.01); A61B 5/7267 (2013.01); G01R 33/4806 (2013.01); G01R 33/4808 (2013.01); G06N 20/00 (2019.01); G16H 30/20 (2018.01); G16H 30/40 (2018.01); G16H 40/67 (2018.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01); A61B 2562/0223 (2013.01); A61N 1/36053 (2013.01); A61N 1/38 (2013.01); A61N 2/006 (2013.01); G16H 20/10 (2018.01); G16H 20/30 (2018.01); G16H 20/40 (2018.01); G16H 20/70 (2018.01)] 17 Claims
OG exemplary drawing
 
1. A system, comprising:
at least one data processor; and
at least one memory storing instructions which, when executed by the at least one data processor, result in operations comprising:
applying a machine learning model to a representation of a first data corresponding to one or more brain signals of a patient, the machine learning model comprising:
a filter configured to generate, based at least on the representation of the first data, a first plurality of latent signals, and
a regression model configured to generate, based at least on a feature of each of the first plurality of latent signals, a treatment outcome prediction,
the machine learning model having been trained by at least optimizing the filter to generate a reduced quantity of latent signals whose feature minimizes an error in the treatment outcome prediction generated by the regression model, the machine learning model having been trained on training data including a second data corresponding to one or more brain signals of a plurality of subjects and the training data further including a treatment outcome associated with each of the plurality of subjects;
determining a distance between the feature of the first plurality of the latent signals associated with the patient and each of one or more clusters, based at least on the feature of the first plurality of latent signals associated with the patient, wherein the one or more clusters are determined by clustering the second data based at least on the feature of each of a second plurality of latent signals generated by the filter based on the second data, and the clustering generating one or more clusters that each correspond to a type of psychiatric disease, wherein a diagnosis for the patient is determined corresponding to one or more types of a psychiatric disease based on the distance; and
determining and providing a treatment plan comprising a treatment modality and/or treatment type based on the treatment outcome prediction and the determined diagnosis for the patient.