Systems and methods to process electronic images to adjust attributes of the electronic images

Patent metadata

Field	Value
Publication number	US-12412316-B2
Application number	US-202217814072-A
Country	US
Kind code	B2
Filing date	Jul 21, 2022
Priority date	May 12, 2021
Publication date	Sep 9, 2025
Grant date	Sep 9, 2025

Abstract

Systems and methods are disclosed for adjusting attributes of whole slide images, including stains therein. A portion of a whole slide image comprised of a plurality of pixels in a first color space and including one or more stains may be received as input. Based on an identified stain type of the stain(s), a machine-learned transformation associated with the stain type may be retrieved and applied to convert an identified subset of the pixels from the first to a second color space specific to the identified stain type. One or more attributes of the stain(s) may be adjusted in the second color space to generate a stain-adjusted subset of pixels, which are then converted back to the first color space using an inverse of the machine-learned transformation. A stain-adjusted portion of the whole slide image including at least the stain-adjusted subset of pixels may be provided as output.

First claim

What is claimed is:

1. A system for processing electronic images to adjust stains, the system comprising:
a data store storing a plurality of machine-learned transformations associated with a plurality of stain types;
a processor; and
a memory coupled to the processor and storing instructions that, when executed by the processor, cause the system to perform operations comprising:
receiving, as input, an image of a slide-mounted tissue sample subjected to overstaining or understaining with a stain during preparation, wherein the image is comprised of a plurality of pixels in a first color space and includes the stain;
determining, from the plurality of machine-learned transformations stored in the data store, a machine-learned transformation associated with a stain type of the stain;
applying the machine-learned transformation to the plurality of pixels to convert the plurality of pixels from the first color space to a second color space, the second color space being specific to the stain type;
correcting for the overstaining or the understaining by adjusting an amount of the stain in the second color space to generate a stain-adjusted plurality of pixels;
converting the stain-adjusted plurality of pixels from the second color space to the first color space using an inverse of the machine-learned transformation; and
outputting a stain-adjusted image including the stain-adjusted plurality of pixels in the first color space.

2. The system of claim 1 , wherein adjusting the amount of the stain in the second color space comprises adjusting the amount of the stain based on a reference image.

3. The system of claim 1 , wherein the second color space comprises at least two channels, the at least two channels including a first channel associated with a brightness of the stain and a second channel associated with an amount of the stain.

4. The system of claim 3 , wherein adjusting the amount of the stain in the second color space comprises adjusting pixel values in the second channel of the second color space.

5. The system of claim 3 , wherein the operations further comprise adjusting the brightness of the stain in the second color space by adjusting pixel values in the first channel of the second color space.

6. The system of claim 1 , wherein adjusting the amount of the stain in the second color space comprises:
providing for display the plurality of pixels in the second color space and one or more graphical user interface (GUI) control elements for adjusting the stain amount;
receiving input associated with at least one of the one or more GUI control elements; and
adjusting the stain amount based on the input.

7. The system of claim 6 , wherein the one or more graphical user interface (GUI) control elements correspond to a channel in the second color space associated with the amount of the stain.

8. The system of claim 1 , wherein adjusting the amount of the stain includes removing the stain from the image.

9. The system of claim 1 , wherein the stain type of the stain is received as input with the image.

10. The system of claim 1 , wherein the operations further comprise identifying the stain type of the stain by:
extracting one or more feature vectors from the image;
providing the one or more feature vectors as input to a trained machine learning system for predicting stain types; and
receiving, as output from the trained machine learning system, the stain type.

11. The system of claim 1 , wherein the operations further comprise identifying a subset of pixels from the plurality of pixels to be transformed, and applying the machine-learned transformation to only the subset of pixels, the subset of pixels including non-background pixels and non-artifact pixels from the plurality of pixels.

12. A method for processing electronic images to adjust stains, the method comprising:
receiving, as input, an image of a slide-mounted tissue sample subjected to overstaining or understaining with a stain during preparation, wherein the image is comprised of a plurality of pixels in a first color space and includes the stain;
determining, from a plurality of stored machine-learned transformations associated with a plurality of stain types, a machine-learned transformation associated with a stain type of the stain;
applying the machine-learned transformation to the plurality of pixels to convert the plurality of pixels from the first color space to a second color space, the second color space being specific to the stain type;
correcting for the overstaining or the understaining by adjusting an amount of the stain in the second color space to generate a stain-adjusted plurality of pixels;
converting the stain-adjusted plurality of pixels from the second color space to the first color space using an inverse of the machine-learned transformation; and
outputting a stain-adjusted image including the stain-adjusted plurality of pixels in the first color space.

13. The method of claim 12 , wherein adjusting the amount of the stain in the second color space comprises adjusting the amount of the stain based on a reference image.

14. The method of claim 12 , wherein the second color space comprises at least two channels, the at least two channels including a first channel associated with a brightness of the stain and a second channel associated with an amount of the stain.

15. The method of claim 14 , wherein adjusting the amount of the stain in the second color space comprises adjusting pixel values in the second channel of the second color space.

16. The method of claim 14 , further comprising adjusting the brightness of the stain in the second color space by adjusting pixel values in the first channel of the second color space.

17. The method of claim 12 , wherein adjusting the amount of the stain in the second color space comprises:
providing for display the plurality of pixels in the second color space and one or more graphical user interface (GUI) control elements for adjusting the stain amount;
receiving input associated with at least one of the one or more GUI control elements; and
adjusting the stain amount based on the input.

18. The method of claim 17 , wherein the one or more graphical user interface (GUI) control elements correspond to a channel in the second color space associated with the amount of the stain.

19. The method of claim 12 , wherein adjusting the amount of the stain includes removing the stain from the image.

20. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform operations for processing electronic images to adjust stains, the operations comprising:
receiving, as input, an image of a slide-mounted tissue sample subjected to overstaining or understaining with a stain during preparation, wherein the image is comprised of a plurality of pixels in a first color space and includes the stain;
determining, from a plurality of stored machine-learned transformations associated with a plurality of stain types, a machine-learned transformation associated with a stain type of the stain;
applying the machine-learned transformation to the plurality of pixels to convert the plurality of pixels from the first color space to a second color space, the second color space being specific to the stain type;
correcting for the overstaining or the understaining by adjusting an amount of the stain in the second color space to generate a stain-adjusted plurality of pixels;
converting the stain-adjusted plurality of pixels from the se

Assignees

• Paige Ai Inc

Inventors

• Alemi Navid
• Kanan Christopher
• Grady Leo

Classifications

• G06T11/10

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T5/00

  Image enhancement or restoration · CPC title

• G06T2210/41

  Medical · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/10024

  Color image · CPC title