US 12,462,550 B2
Systems and methods for identifying cell clusters within images of stained biological samples
Yao Nie, Sunnyvale, CA (US); and Safoora Yousefi, Atlanta, GA (US)
Assigned to VENTANA MEDICAL SYSTEMS, INC., Tucson, AZ (US)
Filed by VENTANA MEDICAL SYSTEMS, INC., Tucson, AZ (US)
Filed on Jan. 26, 2024, as Appl. No. 18/423,572.
Application 18/423,572 is a continuation of application No. 17/213,445, filed on Mar. 26, 2021, granted, now 11,922,681.
Application 17/213,445 is a continuation of application No. PCT/US2019/055558, filed on Oct. 10, 2019.
Claims priority of provisional application 62/830,823, filed on Apr. 8, 2019.
Claims priority of provisional application 62/745,953, filed on Oct. 15, 2018.
Prior Publication US 2024/0161485 A1, May 16, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 10/82 (2022.01); G06F 18/214 (2023.01); G06F 18/231 (2023.01); G06F 18/243 (2023.01); G06V 10/25 (2022.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01); G06V 20/69 (2022.01)
CPC G06V 10/82 (2022.01) [G06F 18/2155 (2023.01); G06F 18/231 (2023.01); G06F 18/243 (2023.01); G06V 10/25 (2022.01); G06V 10/454 (2022.01); G06V 10/764 (2022.01); G06V 20/695 (2022.01); G06V 20/698 (2022.01); G06V 2201/03 (2022.01)] 14 Claims
OG exemplary drawing
 
1. A system for training a cell detection and classification engine, the system comprising: (i) one or more processors, and (ii) one or more memories coupled to the one or more processors, the one or more memories to store computer-executable instructions that, when executed by the one or more processors, cause the system to perform operations comprising:
a. identifying one or more homogeneous clusters of cells within the sample image of the biological specimen stained with the primary stain or stained for the presence of one or more biomarkers, wherein the identifying of the one or more homogeneous clusters of cells comprises:
(i) detecting cells in the sample image using a trained object detection engine, wherein the trained object detection engine comprises first a convolutional neural network adapted to detect cellular features within the sample image, wherein the object detection engine is trained using a dataset comprising a plurality of training images, wherein each training image of the plurality of training images is derived from a training biological specimen stained with a primary stain or stained for the presence of one or more biomarkers, wherein each training image of the plurality of training images does not comprise any pathologist annotations;
(ii) extracting cellular features from one or more layers of the convolutional neural network; and
(iii) clustering the detected cells in the sample image based on the extracted cellular features to provide the one or more homogeneous clusters of detected cells, and
b. training the cell detection and classification engine using pathologist annotations of the one or more identified homogeneous clusters of cells.