Systems and methods for processing images to determine biomarker levels

Patent metadata

Field	Value
Publication number	US-12579645-B2
Application number	US-202318451507-A
Country	US
Kind code	B2
Filing date	Aug 17, 2023
Priority date	Aug 18, 2022
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

Systems and methods are described herein for processing electronic medical images to predict a biomarker's presence, including receiving one or more digital medical images, the one or more digital medical images being of at least one pathology specimen associated with a patient. A machine learning system may determine a biomarker expression level prediction for the one or more digital medical images. The biomarker expression level prediction may be based on a determined transcriptomic score and protein expression score for the one or more digital medical images. A slide overlay indicating a region of tissue on the one or more digital medical images that is most likely to contribute to the slide level biomarker expression prediction may be generated.

First claim

What is claimed is:

1 . A computer-implemented method for processing electronic medical images to predict a biomarker's presence, comprising:
receiving one or more digital medical images, the one or more digital medical images being of at least one pathology specimen associated with a patient; determining a protein expression score for the one or more digital medical images, the protein expression score including an immunohistochemistry (IHC) score for each of the one or more digital medical images; determining a transcriptomic score including a level of Erb-B2 Receptor Tyrosine Kinase 2 (ERBB2) mRNA; determining, by a machine learning system, a biomarker expression level prediction for the one or more digital medical images, the biomarker expression level prediction being based on the transcriptomic score and the protein expression score for the one or more digital medical images, wherein the biomarker expression level prediction is determined to be a true absence of human epidermal growth factor receptor 2 (HER2) expression upon determining that the immunohistochemistry score is IHC-0+, indeterminate, or equivocal-IHC-1+ and that the ERBB2 mRNA level is less than a predetermined value; and generating a slide overlay indicating a region of tissue on the one or more digital medical images that contributes to the biomarker expression level prediction.

2 . The method of claim 1 , further comprising:
determining, salient regions of the received one or more digital medical images prior to determining the biomarker expression level, wherein non-salient image regions are excluded from subsequent processing.

3 . The method of claim 2 , wherein the one or more salient regions correspond to cancerous tissue.

4 . The method of claim 1 , wherein the one or more digital medical images are images of breast tissue stained with hematoxylin and eosin.

5 . The method of claim 1 , wherein the biomarker expression is human epidermal growth factor receptor 2.

6 . The method of claim 1 , wherein the biomarker expression level prediction is performed upon determining that the received one or more slides has a immunohistochemistry (IHC) score of IHC-0 or IHC-1.

7 . The method of claim 1 , wherein the predetermined value of ERBB2 mRNA is 7.6.

8 . The method of claim 1 , wherein generating a slide overlay includes generating a tissue map overlay and/or a heatmap overlay.

9 . A system for processing electronic medical images, the system comprising:
at least one memory storing instructions; and at least one processor configured to execute the instructions to perform operations comprising:
receiving one or more digital medical images, the one or more digital medical images being of at least one pathology specimen associated with a patient;
determining a protein expression score for the one or more digital medical images, the protein expression score including an immunohistochemistry (IHC) score for each of the one or more digital medical images;
determining a transcriptomic score including a level of Erb-B2 Receptor Tyrosine Kinase 2 (ERBB2) mRNA;
determining, by a machine learning system, a biomarker expression level prediction for the one or more digital medical images, the biomarker expression level prediction being based on the transcriptomic score and the protein expression score for the one or more digital medical images, wherein the biomarker expression level prediction is determined to be a true absence of human epidermal growth factor receptor 2 (HER2) expression upon determining that the immunohistochemistry score is IHC-0+, indeterminate, or equivocal-IHC-1+ and that the ERBB2 mRNA level is less than a predetermined value; and
generating a slide overlay indicating a region of tissue on the one or more digital medical images that contributes to the biomarker expression level prediction.

10 . The system of claim 9 , further comprising:
determining, salient regions of the received one or more digital medical images prior to determining the biomarker expression level, wherein non-salient image regions are excluded from subsequent processing.

11 . The system of claim 10 , wherein the one or more salient regions correspond to cancerous tissue.

12 . The system of claim 9 , wherein the biomarker expression level prediction is performed upon determining that the received one or more slides has a immunohistochemistry (IHC) score of IHC-0 or IHC-1.

13 . The system of claim 9 , wherein the one or more digital medical images are images of breast tissue stained with hematoxylin and eosin.

14 . The system of claim 9 , wherein the predetermined value is predetermined value of ERBB2 mRNA is 7.6.

15 . A non-transitory computer-readable medium storing instructions that, when executed by a processor, perform operations processing electronic medical images, the operations comprising:
receiving one or more digital medical images, the one or more digital medical images being of at least one pathology specimen associated with a patient; determining a protein expression score for the one or more digital medical images, the protein expression score including an immunohistochemistry (IHC) score for each of the one or more digital medical images; determining a transcriptomic including a level of Erb-B2 Receptor Tyrosine Kinase 2 (ERBB2) mRNA; determining, by a machine learning system, a biomarker expression level prediction for the one or more digital medical images, the biomarker expression level prediction being based on the transcriptomic score and the protein expression score for the one or more digital medical images, wherein the biomarker expression level prediction is determined to be a true absence of human epidermal growth factor receptor 2 (HER2) expression upon determining that the immunohistochemistry score is IHC-0+, indeterminate, or equivocal-IHC-1+ and that the ERBB2 mRNA level is less than a predetermined value; and generating a slide overlay indicating a region of tissue on the one or more digital medical images that contributes to the biomarker expression level prediction.

16 . The non-transitory computer-readable medium of claim 15 , further comprising:
determining, salient regions of the received one or more digital medical images prior to determining the biomarker expression level, wherein non-salient image regions are excluded from subsequent processing.

17 . The non-transitory computer-readable medium of claim 16 , wherein the one or more salient regions correspond to cancerous tissue.

18 . The non-transitory computer-readable medium of claim 15 , wherein the one or more digital medical images are images of breast tissue stained with hematoxylin and eosin.

19 . The non-transitory computer-readable medium of claim 15 , wherein the predetermined value of ERBB2 mRNA is 7.6.

Assignees

• Paige Ai Inc

Inventors

• Sue Jillian
• Goldfinger Marc
• Rothrock Brandon
• Lee Matthew

Classifications

• G06T2207/30068

  Mammography; Breast · CPC title

• G06T2207/30024

  Cell structures in vitro; Tissue sections in vitro · CPC title

• G06T2207/20221

  Image fusion; Image merging · CPC title

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title

• G06V10/462

  Salient features, e.g. scale invariant feature transforms [SIFT] · CPC title