US 11,948,298 B2
System to collect and identify medical conditions from images and expert knowledge
Abdellatif Abid, Los Angeles, CA (US); Albert Jimenez Sanfiz, Toronto (CA); Adria Romero Lopez, Toronto (CA); Eric T. Jarmain, Toronto (CA); Mohamed Akrout, Toronto (CA); Anirudh Challa, Toronto (CA); Jeremy G. Kawahara, Nanaimo (CA); and Stephen A. Solis-Reyes, London (CA)
Assigned to Triage Technologies Inc., Toronto (CA)
Filed by Triage Technologies Inc., Toronto (CA)
Filed on Sep. 7, 2021, as Appl. No. 17/468,068.
Application 17/468,068 is a continuation of application No. 17/122,329, filed on Dec. 15, 2020.
Application 17/122,329 is a continuation of application No. 16/880,622, filed on May 21, 2020, granted, now 10,878,567, issued on Dec. 29, 2020.
Claims priority of provisional application 62/902,354, filed on Sep. 18, 2019.
Prior Publication US 2021/0407083 A1, Dec. 30, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/00 (2017.01); A61B 5/00 (2006.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC G06T 7/0012 (2013.01) [A61B 5/441 (2013.01); A61B 5/7267 (2013.01); A61B 5/7275 (2013.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06T 2207/20076 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30088 (2013.01); G06T 2207/30168 (2013.01)] 22 Claims
OG exemplary drawing
 
1. A system configured for determining medical conditions from images uploaded to a medical condition classification application that are processed within the medical condition classification application or uploaded to a server, the system comprising:
one or more hardware processors configured by machine-readable instructions to:
store one or more datasets of medical condition images, wherein the medical condition images are associated with one or more medical conditions as determined by human experts;
train a visual classifier model with the one or more datasets of the medical condition images;
receive, via a user interface of the medical condition classification application, one or more uploaded images;
pass the uploaded images through a visual classifier model to determine a context of the uploaded images;
generate an output of the visual classifier model representing predicted probabilities that the uploaded images exhibits one or more medical condition classes;
determine a set of predictions for the medical condition classes;
display the set of predictions that the uploaded images exhibit on a computer display via the user interface;
display information about the set of predictions;
display exemplar medical disease images of the set of predictions or medical disease that are similar to the uploaded images;
generate one or more datasets of validated images of medical conditions by:
generating a queue of images to be presented to the human experts, presenting a second image to a predetermined number of the human experts via the user interface, receiving and storing labels and corresponding probabilities for the second image from human experts, and combining reviews made by the human experts;
remove the second image from the queue; and
store the labels and corresponding probabilities with the second image.