US 12,462,896 B2
Method and system for predicting response to immune anticancer drugs
Donggeun Yoo, Seoul (KR); Chanyoung Ock, Seoul (KR); and Kyunghyun Paeng, Seoul (KR)
Assigned to LUNIT INC., Seoul (KR)
Filed by LUNIT INC., Seoul (KR)
Filed on Oct. 15, 2021, as Appl. No. 17/502,339.
Application 17/502,339 is a continuation of application No. PCT/KR2021/005771, filed on May 7, 2021.
Claims priority of application No. 10-2020-0055483 (KR), filed on May 8, 2020; application No. 10-2021-0054206 (KR), filed on Apr. 27, 2021; and application No. 10-2021-0059537 (KR), filed on May 7, 2021.
Prior Publication US 2022/0036971 A1, Feb. 3, 2022
Int. Cl. G01N 33/48 (2006.01); G06N 3/02 (2006.01); G16B 20/00 (2019.01)
CPC G16B 20/00 (2019.02) [G06N 3/02 (2013.01)] 20 Claims
 
1. A method of predicting response to an immune checkpoint inhibitor, comprising, using at least one computing device:
receiving a first pathology slide image;
detecting one or more target items comprising a cancer area, a stroma area, and immune cells in the first pathology slide image using an artificial neural network model, by:
inputting the first pathology slide image into the artificial neural network model; and
detecting, with the artificial neural network model, the cancer area, the stroma area, and the immune cells in the first pathology slide image, wherein the artificial neural network model is trained to reduce errors between label information for one or more reference target items corresponding to reference pathology slide images and detected output from the reference pathology slide images;
identifying a plurality of patches of a predetermined size in the first pathology slide image, as regions of interest;
determining immune phenotypes of the plurality of patches in the first pathology slide image, based on at least one of a density, a number, or a distribution of immune cells in the cancer area or the stroma area of each of the plurality of patches;
calculating a prediction score indicating whether or not a patient associated with the first pathology slide image responds to an immune checkpoint inhibitor, based on a ratio or a number of patches having an immune phenotype of a specific class among the plurality of patches; and
outputting the prediction score.