Image analysis algorithms using control slides

1. A non-transitory computer-readable medium storing instructions which, when executed by a processor of an image analysis system, cause the image analysis system to perform one or more operations comprising:

obtaining an immunohistochemistry (IHC) image and a control image;

performing a registration operation to align the IHC image and the control image to a single coordinate system;

as a result of the IHC image and the control image being aligned to the single coordinate system based on the registration operation, identifying a distance metric between an image object of the IHC image to a corresponding image object of the control image;

determining that the IHC image and the control image are structurally similar by using the identified distance metric;

determining that the control image includes one or more artifacts that are falsely identifiable as stained cells; and

in response to determining that the control image includes the one or more artifacts, performing an image analysis of a plurality of regions of the IHC image except for one or more artifact regions, wherein the one or more artifact regions of the IHC image correspond to one or more regions of the control image that include the one or more artifacts.

2. The non-transitory computer-readable medium of claim 1 , wherein the IHC image and the control image include tissue sections of a same biological specimen block.

3. The non-transitory computer-readable medium of claim 1 , wherein the image analysis comprises a whole-slide scoring algorithm.

4. The non-transitory computer-readable medium of claim 1 , wherein the IHC image corresponds to a tissue slide stained with a stain of a first color, wherein the one or more artifacts of the control image have a second color, and wherein the first color is substantially similar to the second color.

5. The non-transitory computer-readable medium of claim 1 , wherein the control image corresponds to one of a negative control slide or a positive control slide, and wherein the one or more artifacts of the control image comprise pigments.

6. The non-transitory computer-readable medium of claim 1 , wherein the registration operation comprises transforming the IHC image or the control image using an affine transformation.

7. The non-transitory computer-readable medium of claim 1 , wherein the control image is an image of a negative control slide stained with a counter-stain, and wherein the determining that the control image includes the one or more artifacts comprises determining pixels in the control image fall outside of a color range associated with the counter-stain.

8. The non-transitory computer-readable medium of claim 7 , wherein the color range is either predefined or dynamically computed based on the control image.

9. The non-transitory computer-readable medium of claim 1 , wherein the determining the control image includes the one or more artifacts includes identifying a set of pixels having intensity values that are lower than a predefined threshold.

10. The non-transitory computer-readable medium of claim 1 , wherein:

the control image depicts a first tissue section of a biological sample; and

the IHC image depicts a second tissue section of the biological sample, wherein the first tissue section and the second tissue section are obtained from different slides of a same tissue block.

11. The non-transitory computer-readable medium of claim 1 , wherein instructions further cause the image analysis system to perform one or more operations comprising highlight the one or more artifact regions on a display.

12. The non-transitory computer-readable medium of claim 11 , wherein instructions further cause the image analysis system to perform one or more operations comprising modifying the one or more artifact regions based on a user input.

13. A method comprising:

accessing a field of view (FOV) of an immunohistochemistry (IHC) image of an IHC slide;

accessing a control image of a control slide associated with the IHC slide;

performing a registration operation to align the IHC image and the control image to a single coordinate system;

as a result of the IHC image and the control image being aligned to the single coordinate system based on the registration operation, identifying a distance metric between an image object of the IHC image to a corresponding image object of the control image;

determining that the IHC image and the control image are structurally similar by using the identified distance metric;

detecting artifacts within a region of the control image; and

identifying a presence of artifacts within a region of the FOV that corresponds to the region of the control image that includes the detected artifacts.

14. The method of claim 13 , wherein the control slide and the IHC slide include tissue sections from a same tissue block.

15. The method of claim 13 , wherein the registration operation includes transforming the IHC image or the control image using an affine transformation.

16. The method of claim 13 , wherein the control image is an image of a negative control slide stained with a counter-stain, and wherein the detecting the artifacts comprises determining that pixels in the region of the control image fall outside of a color range associated with the counter-stain.