US 12,402,840 B2
Methods and apparatus to determine developmental progress with artificial intelligence and user input
Brent Vaughan, Portola Valley, CA (US); Clara Lajonchere, Los Angeles, CA (US); Dennis Wall, Palo Alto, CA (US); Jay Hack, Ann Arbor, MI (US); and Charlie Hack, New York, NY (US)
Assigned to Cognoa, Inc., Palo Alto, CA (US)
Filed by Cognoa, Inc., Palo Alto, CA (US)
Filed on Nov. 18, 2020, as Appl. No. 16/951,915.
Application 16/951,915 is a continuation of application No. 15/234,814, filed on Aug. 11, 2016, granted, now 10,874,355.
Claims priority of provisional application 62/203,777, filed on Aug. 11, 2015.
Prior Publication US 2021/0068766 A1, Mar. 11, 2021
Int. Cl. A61B 5/00 (2006.01); A61B 5/16 (2006.01); G16H 10/20 (2018.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01); G16H 50/70 (2018.01)
CPC A61B 5/7275 (2013.01) [A61B 5/00 (2013.01); A61B 5/16 (2013.01); G16H 10/20 (2018.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01); G16H 50/70 (2018.01); A61B 5/168 (2013.01)] 37 Claims
OG exemplary drawing
 
1. An apparatus for treating a subject for autism spectrum disorder, said apparatus comprising:
a processor; and
a non-transitory computer readable storage medium including instructions that, when executed by said processor, cause said processor to:
(a) receive a first input corresponding to a first feature from a set of features, wherein said set of features corresponds to one or more clinical characteristics of said subject related to said autism spectrum disorder and a related developmental disorder;
(b) determine, based at least in part on said first input corresponding to said first feature, that said subject has a greater risk of (i) said autism spectrum disorder than (ii) said related developmental disorder;
(c) in response to determining said subject has said greater risk of said autism spectrum disorder than said related developmental disorder, identify a second feature from said set of features, wherein said second feature is most predictive of said autism spectrum disorder, wherein said second feature is identified based at least in part on a first expectation value of said second feature being greater than a second expectation value of a third feature, wherein:
(i) said first expectation value is based at least in part on a plurality of first predictive utilities of each of a plurality of first possible inputs for said second feature and a plurality of first probabilities of occurrence of each of said plurality of first possible inputs for said second feature, wherein each of said plurality of first probabilities of occurrence is determined based at least in part on said first feature and a first constant coefficient that is determined by a trained assessment model, and
(ii) said second expectation value is based at least in part on a plurality of second predictive utilities of each of a plurality of second possible inputs for said third feature and a plurality of second probabilities of occurrence of each of said plurality of second possible inputs for said third feature, wherein each of said plurality of second probabilities of occurrence is determined based at least in part on said first feature and a second constant coefficient that is determined by said trained assessment model;
(d) receive a second input corresponding to said second feature that is most predictive of said autism spectrum disorder;
(e) determine, based at least in part on said first input corresponding to said first feature and said second input corresponding to said second feature, that said subject has said greater risk of said autism spectrum disorder than said related developmental disorder; and
(f) responsive to said determining in (e), provide a therapy to said subject to treat said autism spectrum disorder, wherein said therapy comprises applied behavioral analysis (ABA) or speech therapy.