Detection of labels in biomedical images

Patent metadata

Field	Value
Publication number	US-12555672-B2
Application number	US-202218558708-A
Country	US
Kind code	B2
Filing date	May 2, 2022
Priority date	May 3, 2021
Publication date	Feb 17, 2026
Grant date	Feb 17, 2026

Abstract

Presented herein are systems and methods for detecting labels in biomedical images. A computing system having one or more processors coupled with memory may identify, from a data source, a biomedical image having a first plurality of pixels in a first color representation. The computing system may convert the first plurality of pixels from the first color representation to a second color representation to generate a second plurality of pixels. The computing system may identify, from the second plurality of pixels, a subset of pixels having a color value satisfying a threshold value. The computing system may detect the biomedical image as having at least one label based at least on a number of pixels in the subset of pixels satisfying a threshold count. The computing system may store, in one or more data structures, an indication for the biomedical image as having the at least one label.

First claim

What is claimed is:

1 . A system for detecting labels in biomedical images, comprising:
a computing system having one or more processors coupled with memory, configured to: identify a first biomedical image having a first plurality of pixels in a first color representation, the first biomedical image a whole slide image (WSI) of a sample tissue on a slide; convert the first plurality of pixels from the first color representation to a second color representation to generate a second plurality of pixels; identify, from the second plurality of pixels, a first subset of pixels having a color value satisfying a threshold value; detect the first biomedical image as having at least one label based at least on a first number of pixels in the first subset of pixels satisfying a threshold count; store, in one or more first data structures, a first indication for the first biomedical image as having the at least one label; identify a second biomedical image as lacking any label based at least on a second number of pixels in a second subset of pixels that satisfy the threshold value not satisfying the threshold count; store, in one or more second data structures, a second indication for the second biomedical image as lacking any label; and permit transfer of a second biomedical image, responsive to identifying the second biomedical image as lacking any label.

2 . The system of claim 1 , wherein the computing system is further configured to restrict transfer of the first biomedical image, responsive to detecting the first biomedical image as having the at least one label.

3 . The system of claim 1 , wherein the computing system is further configured to:
determine that the first biomedical image is not maintained separately from the at least one label associated with the first biomedical image convert the first plurality of pixels from the first color representation to the second color representation, responsive to determining that the first biomedical image is not maintained separately from the at least one label.

4 . The system of claim 1 , wherein the computing system is further configured to identify, based at least on a data source from which the first biomedical image is received, at least one of the second color representation, the threshold value, or the threshold count.

5 . The system of claim 1 , wherein the computing system is further configured to:
determine that the second biomedical image is separately maintained from a second label associated with the second biomedical image; and
permit transfer of a first file corresponding to the second biomedical image while removing a second file corresponding to the second label.

6 . The system of claim 1 , wherein the computing system is further configured to determine whether to provide the first biomedical image in response to a request, based at least on the first indication for the first biomedical image.

7 . The system of claim 1 , wherein the computing system is further configured to select, from a plurality of biomedical images, the first biomedical image based at least on a magnification factor for the first biomedical image from which to detect the at least one label.

8 . The system of claim 1 , wherein the WSI is acquired in accordance with microscopy, wherein the sample tissue is obtained from an anatomical location of a subject, wherein the slide has a portion corresponding to the at least one label.

9 . A method of detecting labels in biomedical images, comprising:
identifying, by a computing system, a first biomedical image having a first plurality of pixels in a first color representation, the first biomedical image a whole slide image (WSI) of a sample tissue on a slide; converting, by the computing system, the first plurality of pixels from the first color representation to a second color representation to generate a second plurality of pixels; identifying, by the computing system, from the second plurality of pixels, a first subset of pixels having a color value satisfying a threshold value; detecting, by the computing system, the first biomedical image as having at least one label based at least on a first number of pixels in the first subset of pixels satisfying a threshold count; storing, by the computing system, in one or more first data structures, a first indication for the first biomedical image as having the at least one label; identifying, by the computing system, a second biomedical image as lacking any label based at least on a second number of pixels in a second subset of pixels that satisfy the threshold value not satisfying the threshold count; storing, by the computing system, in one or more second data structures, a second indication for the second biomedical image as lacking any label; and permitting, by the computing system, transfer of a second biomedical image, responsive to identifying the second biomedical image as lacking any label.

10 . The method of claim 9 , further comprising restricting, by the computing system, transfer of the first biomedical image, responsive to detecting the first biomedical image as having the at least one label.

11 . The method of claim 9 , further comprising determining, by the computing system, that the first biomedical image is not maintained separately from the at least one label associated with the first biomedical image, and
wherein converting further comprises converting the first plurality of pixels from the first color representation to the second color representation, responsive to determining that the first biomedical image is not maintained separately from the at least one label.

12 . The method of claim 9 , further comprising identifying, by the computing system, based at least on a data source from which the first biomedical image is received, at least one of the second color representation, the threshold value, or the threshold count.

13 . The method of claim 9 , further comprising:
determining, by the computing system, that the second biomedical image is separately maintained from a second label associated with the second biomedical image; and wherein permitting the transfer further comprises permitting the, transfer of a first file corresponding to the second biomedical image while removing a second file corresponding to the second label.

14 . The method of claim 9 , further comprising determining, by the computing system, whether to provide the first biomedical image in response to a request, based at least on the first indication for the first biomedical image.

15 . The method of claim 9 , wherein identifying the first biomedical image further comprises selecting, from a plurality of biomedical images, the first biomedical image based at least on a magnification factor for the first biomedical image from which to detect the at least one label.

16 . The method of claim 9 , wherein the WSI is acquired in accordance with microscopy, wherein the sample tissue obtained from a subject, wherein the slide has a portion corresponding to the at least one label.

Assignees

• Memorial Sloan Kettering Cancer Center
• Sloan Kettering Inst Cancer Res
• Memorial Hospital For Cancer And Allied Diseases

Inventors

• Geneslaw Luke
• Fuchs Thomas
• Yarlagadda Dig Vijay Kumar

Classifications

• G06T7/0012

  Biomedical image inspection · CPC title

• G06V2201/10

  Recognition assisted with metadata · CPC title

• G06V10/255

  Detecting or recognising potential candidate objects based on visual cues, e.g. shapes · CPC title

• G06V2201/03

  Recognition of patterns in medical or anatomical images · CPC title

• G06V10/56

  relating to colour · CPC title