Detecting and visualizing correlations between measured correlation values and correlation reference values of a pathway

• US 9,990,713 B2
• Filed: 06/09/2016
• Issued: 06/05/2018
• Est. Priority Date: 06/09/2016
• Status: Active Grant

• Current Assignee: Definiens AG (Astrazeneca PLC)
• Original Assignee: Definiens AG (Astrazeneca PLC)
• Inventors: Schoenmeyer, Ralf, Althammer, Sonja, Schmidt, Guenter
• Primary Examiner(s): Woldemariam, Aklilu
• Application Number:
  US15/178,002
• Publication Number:
  US 20170358074A1
• Time in Patent Office: 726 Days
• Field of Search: 382128, 382133, 382173, 382224, 382225, 382226, 382227, 382158, 382195, 382305, 345629, 345428, 356300
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/10 Segmentation; Edge detection motion-based segmentation G06T7/215
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/32 using correlation-based methods
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
• CPC Class Codes: G16B 5/00 ICT specially adapted for modelling or simulations in systems biology, e.g. gene-regulatory networks, protein interaction networks or metabolic networks
• CPC Class Codes: G16B 99/00 Subject matter not provided for in other groups of this subclass
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

An analysis and visualization system analyzes a digital image of a tissue sample. In the sample, cells of a first type are stained in a first way, and cells of a second type are stained in a second way. The system segments the high-resolution image into first and second objects representing cells of the first and second types, respectively. The system also identifies a region of interest, and divides it into tiles. The system generates, for each tile, a first value and a second value. The first and second values for a tile are indicative of densities of the first and second objects in the tile. From the values, a measured correlation coefficient (CC) value is determined. The system compares the measured CC value to a reference CC value, thereby obtaining a correspondence value. The system then displays the image region along with a visualization of the correspondence value.

Claims

1. A method involving a correlation value determining and visualization system, the method comprising:

(a) segmenting a digital image of a tissue slice of a cancer patient into first image objects representing immune cells and second image objects representing cancer cells, wherein the tissue slice was stained with a first antibody that is specific to a first protein present in immune cells, and wherein the tissue slice was stained with a second antibody that is specific to a second protein present in cancer cells;

(b) identifying a reference correlation value from among a plurality of reference correlation values indicative of protein-protein interactions in biochemical pathways, wherein the reference correlation value is indicative of a correlation between immune cells and cancer cells;

(c) identifying an image region in the digital image, wherein the image region is divided into a set of tiles;

(d) generating, for each tile, a first value and a second value, wherein the first value is indicative of a density of the first image objects in the tile, and wherein the second value is indicative of a density of the second image objects in the tile;

(e) determining, for the image region, a measured correlation value that is indicative of a relationship between the first value and the second value;

(f) comparing the measured correlation value determined in (e) with the reference correlation value identified in (b) and thereby generating a correspondence value; and

(g) displaying the image region on a display of the system along with an associated visual indication representing the correspondence value determined in (f), wherein (a) through (g) are performed by the system.

2. The method of claim 1, wherein the system stores a plurality of reference correlation values, wherein the identifying of (b) involves receiving user input into the system, wherein the system then uses the user input to identify one of the plurality of reference correlation values to be the reference correlation value identified in (b).

3. The method of claim 1, wherein the identifying of (b) further involves displaying a representation of a biochemical pathway on the display of the system.

4. The method of claim 1, wherein the identifying of (b) further involves displaying a representation of a biochemical pathway on the display of the system, wherein the representation includes a plurality of user-selectable icons.

5. The method of claim 1, wherein the identifying of (b) further involve displaying to the user on the display of the system an indication of each pathway of a plurality of pathways, and soliciting the user to select one of the pathways.

6. The method of claim 1, wherein the identifying of (b) involves receiving user input into the system, wherein the user input is the reference correlation value.

7. The method of claim 1, wherein the identifying of (c) involves receiving user input into the system, wherein the system then uses the user input to identify the image region in the digital image.

8. The method of claim 1, wherein the identifying of (c) involves dividing the digital image into a two-dimensional array of identically-shaped image regions, wherein the image region identified in (c) is one of these identically-shaped image regions.

9. The method of claim 1, wherein the generating of (d), for each tile, involves counting the first image objects in the tile thereby generating a first count and counting the second image objects in the tile thereby generating a second count, wherein the first count is the first value, and wherein the second count is the second value.

10. The method of claim 1, wherein the determining of (e) is a determination of a Spearman'"'"'s rank correlation coefficient.

11. The method of claim 1, wherein the correspondence value determined in (f) is indicative of the absolute value of a difference between the measured correlation value determined in (e) and the reference correlation value identified in (b).

12. The method of claim 1, wherein the visual indication of (g) is a shading of the image region, wherein a degree of the shading indicates the magnitude of the correspondence value.

13. The method of claim 12, wherein the visual indication of (g) further comprises a shading key, wherein the shading key indicates, for each possible shade of the image region, a corresponding correspondence value.

21. The method of claim 1, wherein the reference correlation value quantifies a strength of the relationship between the first image objects and the second image objects.

14. A method involving a correlation value determining and visualization system, the method comprising:

(a) receiving a digital image of a tissue sample of a cancer patient into the system, wherein the tissue sample includes first image objects stained with a first antibody stain, and wherein the tissue sample also includes second image objects stained with a second antibody stain;

(b) identifying in the digital image the first image objects;

(c) identifying in the digital image the second image objects;

(d) identifying a reference correlation value from among a plurality of reference correlation values indicative of protein-protein interactions in biochemical pathways, wherein the reference correlation value is indicative of a correlation between the first image objects and the second image objects;

(e) identifying an image region in the digital image, wherein the image region is divided into a set of tiles;

(f) generating, for each tile of the image region, a first value and a second value, wherein the first value is indicative of a density of the first image objects in the tile, and wherein the second value is indicative of a density of the second image objects in the tile;

(g) determining a measured correlation value indicative of a relationship between the first value and the second value;

(h) comparing the measured correlation value determined in (g) with the reference correlation value identified in (d) thereby generating a correspondence value; and

(i) displaying the image region on a display of the system along with an associated visual indication representing the correspondence value determined in (h), wherein (a) through (h) are performed by the system.

15. The method of claim 14, wherein the receiving of (a) is a storing of a digital image file into the system, and wherein the correspondence value generated in (h) is displayed in (i) as a visual indication of a difference between the measured correlation value determined in (g) and the reference correlation value identified in (d).

16. The method of claim 14, wherein the digital image represents a region, wherein the identifying of (e) is a dividing of the region of the digital image into a two-dimensional array of identically shaped image regions, wherein the image region identified in (e) is one of the image regions of the two-dimensional array.

17. The method of claim 14, wherein the identifying of the reference correlation value of (d) involves displaying a representation of a biochemical pathway on the display of the system, wherein the representation includes a plurality of user-selectable icons, and wherein the identifying of the correlation value of (d) further involves receiving user pathway icon selection information into the system.

18. A method involving a correlation value determining and visualization system, the method comprising:

(a) storing digital image information, of a tissue sample of a cancer patient on the system, wherein the tissue sample includes first image objects stained with a first stain, and wherein the tissue sample also includes second image objects stained with a second stain;

(b) storing a plurality of reference correlation values on the system that indicate how specified proteins correlate with other proteins in biochemical pathways, wherein one of the plurality of reference correlation values is indicative of a correlation between objects stained with the first stain and objects stained with the second stain;

(c) storing a plurality of measured correlation values on the system, wherein one of the measured correlation values is indicative of a correlation between the first image objects and the second image objects, wherein said one measured correlation value was generated by the system from the digital image information;

(d) receiving user input information into the system, wherein the user input information selects a biochemical pathway; and

(e) in response to the receiving in (d) causing a visual indication of a correspondence value to be displayed on a display of the system, wherein the correspondence value is indicative of a comparison of said one measured correlation value with said one of the plurality of reference correlation values, and wherein (a) through (e) are performed by the system.

19. The method of claim 18, further comprising:

(f) in response to the receiving of user input information in (d) causing a representation of the selected biochemical pathway to be displayed on the system, wherein the representation of the selected biochemical pathway includes a first pathway node icon and a second pathway node icon, wherein the first pathway node icon represents a first protein, wherein the second pathway node represents a second protein, and wherein the one reference correlation value in (e) is a reference correlation value of the first protein with respect to the second protein.

20. The method of claim 19, wherein the first pathway node icon is a user selectable icon, wherein the second pathway node icon is a selectable icon, wherein the user input information received in (d) includes information indicating that the first pathway node icon has been selected, and wherein the user input information received in (d) further includes information indicating that the second pathway node icon has been selected.