Gleason Grading by Segmenting and Combining Co-Registered Images of Differently Stained Tissue Slices

• US 20140169654A1
• Filed: 12/19/2012
• Published: 06/19/2014
• Est. Priority Date: 12/19/2012
• Status: Active Grant

• Current Assignee: Definiens AG (Astrazeneca PLC)
• Original Assignee: Definiens AG (Astrazeneca PLC)
• Inventors: Athelogou, Maria, Binnig, Gerd, Kirchner, Thomas, Hessel, Harald
• Granted Patent: US 8,879,819 B2
• Time in Patent Office: Days
• US Class Current: 382/133
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 2207/30081 Prostate
• CPC Class Codes: G06T 2207/30242 Counting objects in image
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/174 involving the use of two or more images
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts

Abstract

An improved histopathological score is obtained by generating image objects from images of tissue containing stained epithelial cells. First objects are generated that correspond to basal cells stained with a first stain, such as p63. Second objects are generated that correspond to luminal cells stained with a second stain, such as CK18. If the same tissue is not stained with both stains, then the images of differently stained tissue are co-registered. Third objects are defined to include only those second objects that have more than a minimum separation from any first object. A scoring region includes the third objects, and the histopathological score is determined based on tissue that falls within the scoring region. For example, a Gleason score of prostate tissue is determined by classifying tissue patterns in the scoring region. Alternatively, a Gleason pattern is assigned by counting the number of third objects that possess a predetermined form.

Claims

1. A method comprising:

generating first objects from a first digital image of a first tissue slice, wherein the first tissue slice has been stained with a first stain that stains basal epithelial cells;

generating second objects from a second digital image of a second tissue slice, wherein the second tissue slice has been stained with a second stain that stains luminal epithelial cells;

co-registering the first digital image with the second digital image;

defining third objects that include only those second objects that have more than a minimum separation from any of the first objects;

defining a scoring region that includes the third objects; and

determining a histopathological score in the scoring region.

2. The method of claim 1, wherein the histopathological score is determined based on tissue stained with the second stain.

3. The method of claim 1, wherein the first stain is transcription factor p63, and wherein the transcription factor p63 comprises a protein-specific antibody and an attached dye.

4. The method of claim 1, wherein the first stain stains basal epithelial cells by staining nuclei inside the basal epithelial cells.

5. The method of claim 1, wherein the first stain is transcription factor p63 and the second stain is cytokeratin 18 (CK18).

6. The method of claim 1, further comprising:

co-registering the second digital image and a third digital image, wherein the third digital image is of a third tissue slice stained with a third stain;

locating the scoring region on the third digital image; and

generating fourth objects in the scoring region in the third digital image that have been stained with the third stain, wherein the histopathological score is determined based on tissue stained with the third stain.

7. The method of claim 6, wherein the third stain is hematoxylin and eosin (H&

E).

8. The method of claim 1, wherein the histopathological score is taken from the group consisting of:

a Gleason score, an Allred score, an Elston-Ellis score, an Her2 score, and a HercepTest score.

9. The method of claim 1, wherein the minimum separation is 150 microns.

10. The method of claim 1, wherein the defining the third objects involves growing the second objects by the minimum separation and excluding from the third objects those second objects that have grown into any of the first objects.

11. The method of claim 1, wherein the third objects are a subset of the second objects.

12. The method of claim 1, wherein the first tissue slice originated adjacent to the second tissue slice.

13. The method of claim 1, wherein the first tissue slice and the second tissue slice where adjacent to one another in a tissue sample.

14. The method of claim 13, wherein the tissue sample is a needle tissue sample.

15. The method of claim 1, wherein the co-registering is performed before the identifying the first objects.

16. A method comprising:

segmenting a first digital image to generate first objects corresponding to tissue stained with a first stain, wherein the first stain stains basal membranes of the tissue;

segmenting a second digital image to generate second objects corresponding to tissue stained with a second stain, wherein the second stain stains epithelial cells in the tissue;

co-registering the first digital image and the second digital image; and

displaying the second digital image on a graphical user interface with a subset of the second objects identified, wherein the subset of the second objects includes only those second objects that have at least a minimum separation from any of the first objects.

17. The method of claim 16, further comprising:

determining a Gleason score based on the subset of the second objects.

18. The method of claim 17, wherein the determining the Gleason score is performed by counting a number of the second objects present in the subset of the second objects.

19. The method of claim 16, further comprising:

classifying an area of the tissue that includes the subset of the second objects as belonging to a Gleason pattern.

20. The method of claim 16, wherein the second stain is cytokeratin 18 (CK18).

21. The method of claim 16, wherein the minimum separation is a number of pixels.

22. A method comprising:

identifying first objects in a digital image of a tissue slice, wherein the tissue slice has been stained with a first stain that stains basal epithelial cells and with a second stain that stains luminal epithelial cells;

identifying second objects in the digital image of the tissue slice;

defining third objects that include only those second objects that have more than a minimum separation from any of the first objects;

counting a number of the third objects that possess a predetermined form; and

determining a histopathological score based in part on the number of the third objects that possess the predetermined form.

23. The method of claim 22, wherein the histopathological score is taken from the group consisting of:

a Gleason score, an Allred score, an Elston-Ellis score, an Her2 score, and a HercepTest score.

24. The method of claim 22, wherein the first stain is cytokeratin 14 (CK14) and the second stain is cytokeratin 18 (CK18).

25. The method of claim 22, wherein the predetermined form is defined by an object feature taken from the group consisting of:

length, width, area, asymmetry, elliptic fit, roundness, and number of branches.