EVALUATION SYSTEM, EVALUATION METHOD, AND RECORDING MEDIUM

• US 20090116724A1
• Filed: 11/05/2008
• Published: 05/07/2009
• Est. Priority Date: 11/06/2007
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Saito, Akira, Okajima, Kenji, Yamashita, Yoshiko
• Granted Patent: US 8,111,897 B2
• Time in Patent Office: Days
• US Class Current: 382/133
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts

Abstract

An evaluation system includes a capture unit for capturing an image of a living tissue in which HER2 protein and cell nucleuses are dyed, a discrimination unit for identifying a cell membrane from the image of the living tissue based on dyed cell nucleuses within the image of the living tissue captured by the capture unit to discriminate a dyed state of the cell membrane, and an evaluation unit for evaluating development of the HER2 protein based on a discrimination result by the discrimination unit.

Claims

1. An evaluation system comprising:

a capture unit that captures an image of a living tissue in which HER2 protein and cell nucleuses are dyed;

a discrimination unit that identifies a cell membrane from the image of the living tissue based on dyed cell nucleuses within the image of the living tissue captured by said capture unit to discriminate a dyed state of the cell membrane; and

an evaluation unit that evaluates development of the HER2 protein based on a discrimination result by said discrimination unit.

2. The evaluation system according to claim 1, wherein said discrimination unit includes:

a dyeing degree discrimination unit that discriminates degree of dyeing of the HER2 protein for each pixel that represents dyed HER2 protein within the image of the living tissue;

a detection unit that detects the dyed cell nucleus from the image of the living tissue; and

a cell membrane dyeing degree discrimination unit that identifies the cell membrane from the image of the living tissue based on the cell nucleus detected in said detection unit, and discriminates the dyed state of the cell membrane based on the degree of dyeing in a pixel corresponding to the cell membrane in each pixel whose degree of dyeing is discriminated in said dyeing degree discrimination unit.

3. The evaluation system according to claim 2, wherein:

said capture unit further captures information about a tumorous region in the image of the living tissue;

said discrimination unit further includes an identification unit that identifies the tumorous region from the image of the living tissue based on the information on the tumorous region captured by said capture unit;

said dyeing degree discrimination unit discriminates the degree of dyeing of the HER2 protein for each pixel that represents the dyed HER2 protein within the tumorous region identified by said identification unit; and

said detection unit detects the dyed cell nucleus from the tumorous region identified by said identification unit.

4. The evaluation system according to claim 2, wherein:

said detection unit detects a cell nucleus of a tumorous cell from the image of the living tissue; and

said cell membrane dyeing degree discrimination unit identifies the cell membrane of the tumorous cell from the image of the living tissue based on the cell nucleus of the tumorous cell detected by said detection unit, and discriminates the dyed state of the cell membrane of the tumorous cell based on the degree of dyeing in a pixel corresponding to the cell membrane of the tumorous cell in each pixel whose degree of dyeing is discriminated by said dyeing degree discrimination unit.

5. The evaluation system according to claim 3, wherein said discrimination unit further includes:

a dyed region proportion determination unit that extracts a dyed region that has the HER2 protein from the tumorous region, and determines a proportion of the dyed region that has the HER2 protein within the tumorous region; and

a strong-dyeing proportion determination unit that determines a proportion at which the tumorous region is occupied by strongly dyed pixels whose degree of dyeing is equal to or more than a predefined reference value based on the pixels whose degree of dyeing is discriminated by said dyeing degree discrimination unit,

wherein said detection unit detects the cell nucleus of the tumorous cell from the tumorous region when the proportion of the dyed region that has the HER2 protein is equal to or more than a predefined reference dyeing value within the tumorous region, and when the proportion of the strongly dyed pixels is less than a predefined reference strong-dyeing value within the tumorous region.

6. The evaluation system according to claim 4, wherein said discrimination unit further includes:

a dyed region proportion determination unit that extracts a dyed region that has the HER2 protein from the image of the living tissue, and determines a proportion of the dyed region that has the HER2 protein within the image of the living tissue; and

a strong dyeing proportion determination unit that determines a proportion at which the image of the living tissue is occupied by strongly dyed pixels whose degree of dyeing is equal to or more than a predefined reference value, based on the pixels whose degree of dyeing is discriminated by said dyeing degree discrimination unit,

wherein said detection unit detects the cell nucleus of the tumorous cell from the image of the living tissue when the proportion of the dyed region that has the HER2 protein is equal to or more than a predefined reference dyeing value within the image of the living tissue, and when the proportion of the strongly dyed pixels is less than a predefined reference strong-dyeing value within the image of the living tissue.

7. The evaluation system according to claim 1, wherein said evaluation unit further outputs display data representative of an image which displays the dyed state of the cell membrane discriminated by said discrimination unit on the image of the living tissue.

8. An evaluation system comprising:

capturing means for capturing an image of a living tissue in which HER2 protein and cell nucleuses are dyed;

discriminating means for identifying a cell membrane from the image of the living tissue based on dyed cell nucleuses within the image of the living tissue captured by said capturing means, and discriminating a dyed state of the cell membrane; and

evaluating means for evaluating development of the HER2 protein based on a discrimination result by said discriminating means.

9. An evaluation method comprising:

capturing an image of a living tissue in which HER2 protein and cell nucleuses are dyed;

discriminating a dyed state of a cell membrane by identifying the cell membrane from the image of the living tissue based on dyed cell nucleuses within the captured image of the living tissue; and

evaluating development of the HER2 protein based on the result of the discrimination.

10. The evaluation method according to claim 9, wherein said discriminating includes:

discriminating degree of dyeing of the HER2 protein for each pixel representative of dyed HER2 protein within the image of the living tissue;

detecting the dyed cell nucleus from the image of the living tissue;

identifying the cell membrane from the image of the living tissue based on the cell nucleus; and

discriminating the dyed state of the cell membrane based on the degree of dyeing in a pixel corresponding to the cell membrane in each pixel whose degree of dyeing is discriminated.

11. The evaluation method according to claim 10, further comprising:

capturing information on a tumorous region in the image of the living tissue; and

identifying the tumorous region from the image of the living tissue based on the information on the tumorous region,

wherein said discriminating the degree of dyeing includes discriminating the degree of dyeing of the HER2 protein for each pixel that represents the dyed HER2 protein within the tumorous region, and

said detecting includes detecting the dyed cell nucleus from the tumorous region.

12. The evaluation method according to claim 10 wherein:

said detecting includes detecting a cell nucleus of a tumorous cell from the image of the living tissue; and

said discriminating the cell membrane degree of dyeing includes identifying the cell membrane of the tumorous cell from the image of the living tissue based on the cell nucleus of the tumorous cell, and discriminating the dyed state of the cell membrane of the tumorous cell based on the degree of dyeing in a pixel corresponding to the cell membrane of the tumorous cell in each pixel whose degree of dyeing is discriminated.

13. The evaluation method according to claim 11, further comprising:

extracting a dyed region that has the HER2 protein from the tumorous region to determine a proportion of the dyed region of the HER2 protein within the tumorous region; and

discriminating a proportion at which the tumorous region is occupied by strongly dyed pixels whose degree of dyeing is equal to or more than a predefined reference value based on the pixels whose degree of dyeing is discriminated,

wherein said detecting includes detecting the cell nucleus from the tumorous region when the proportion of the dyed region that has the HER2 protein is equal to or more than a predefined reference dyeing value within the tumorous region, and when the proportion of the strongly dyed pixels is less than a predefined reference strong-dyeing value within the tumorous region.

14. The evaluation method according to claim 12, further comprising:

extracting a dyed region that has the HER2 protein from the image of the living tissue to determine a proportion of the dyed region that has the HER2 protein within the image of the living tissue; and

discriminating a proportion at which the image of the living tissue is occupied by strongly dyed pixels whose degree of dyeing is equal to or more than a predefined reference value, based on pixels whose degree of dyeing is discriminated,

wherein said detecting includes detecting the cell nucleus of the tumorous cell from the image of the living tissue when the proportion of the dyed region that has the HER2 protein is equal to or more than a predefined reference dyeing value within the image of the living tissue, and when the proportion of the strongly dyed pixels is less than a predefined reference strong-dyeing value within the image of the living tissue.

15. The evaluation method according to claim 9, further comprising outputting display data representative of an image which displays the dyed state of the cell membrane on the image of the living tissue.

16. A computer readable recording medium in which an evaluation program is recorded, said evaluation program causing a computer to execute:

a capture procedure for capturing an image of a living tissue in which HER2 protein and cell nucleuses are dyed;

a discrimination procedure for identifying a cell membrane from the image of the living tissue based on dyed cell nucleuses within the captured image of the living tissue to discriminate a dyed state of the cell membrane; and

an evaluation procedure for evaluating development of the HER2 protein based on the result of the discrimination.

17. The recording medium according to claim 16, wherein said discrimination procedure includes:

a dyeing degree discrimination procedure for discriminating degree of dyeing of the HER2 protein for each pixel representative of dyed HER2 protein within the image of the living tissue;

a detection procedure for detecting the dyed cell nucleus from the image of the living tissue; and

a cell membrane dyeing degree discrimination procedure for identifying the cell membrane from the image of the living tissue based on the cell nucleus, and discriminating the dyed state of the cell membrane based on the degree of dyeing in a pixel corresponding to the cell membrane in each pixel whose degree of dyeing is discriminated.