Computer-implemented process on an image of a biological sample

Patent metadata

Field	Value
Publication number	US-11756194-B2
Application number	US-201917258307-A
Country	US
Kind code	B2
Filing date	Jun 27, 2019
Priority date	Jul 11, 2018
Publication date	Sep 12, 2023
Grant date	Sep 12, 2023

Abstract

Computer-implemented process on an image of a biological sample The present invention relates to a computer-implemented process to automatically analyze a digital image ( 103 ) of abiological sample ( 101 ). The process involves a change ( 203 ) from a first color space to a second color space. Then, fits are performed taking into account several dimensions of the second color space to classify pixels.

First claim

The invention claimed is:

1. A process for classifying pixels of a digital image of a biological sample and comprising the steps of:
(a) receiving a first digital image representing the biological sample, in order to obtain a value X 1 associated with each pixel of the first digital image, and obtaining at least one other value X i for each pixels from at least one of the first digital image and a further digital image, i being an integer such that 2≤i≤n and n being an integer such than n≥2;
(b) changing of space with the matrix multiplication
I=A×X

to obtain a second digital image, X being a matrix of n lines and 1 column including the values X 1 to X n , I being a matrix of m lines and 1 column including values I 1 to I m , m being an integer equal to 2 or higher, I 1 to I m being values associated with each pixel of the second digital image, A being a matrix of m lines and n columns;
(c) determining a classification condition based on the values in I 1 and on the values in I j of at least some pixels in the second digital image with j>1 and j≤m; and
(d) classifying pixels of the second digital image that fulfill the classification condition in a first class;
wherein step (c) comprises:
determining a set of one or more distributions, the set comprising at least one distribution of a number of pixels in the second digital image as function of I 1 and the set comprising at least one distribution of a number of pixels in the second digital image as function of I j with j>1 and j≤m,
fitting the one or more distributions of the set to determine at least one fitting parameter, and
determining the classification condition from the at least one fitting parameter.

2. The process according to claim 1 , wherein the set of one or more distributions comprises a distribution that is function of I 1 and function of I j with j≥1 and j≤m.

3. The process according to claim 1 , wherein the set of one or more distributions comprises a first distribution that is function of I 1 and a second distribution that is function of I j with j>1 and j≤m.

4. The process according to claim 1 , wherein the classification condition for a considered pixel relates to the value of the considered pixel in I 1 and in the at least one I j with j>1 and j≤m.

5. The process according to claim 1 , wherein an empirical calibration parameter is used in the determination of the classification condition, said empirical calibration parameter having been determined by showing calibration biological samples, prepared in the same way as the biological sample, to biological practitioners.

6. The process according to claim 1 , further comprising a determination of the matrix A from a principal component analysis from an independent component analysis, or from a factorial analysis on the matrix X.

7. The process according to claim 1 , wherein n=3, m=3, X 1 is a red value R in the first digital image, X 2 is a green value G in the first digital image, and X 3 is a blue value B in the first digital image, and wherein

A
=

[

1
3

1
3

1
3

1
2

0

-

1
2

-

1
4

1
2

-

1
4

]

.

8. The process according to claim 1 , wherein the first digital image shows at least part of the biological sample stained according to a first staining technique.

9. The process according to claim 8 , wherein the further digital image shows at least part of the biological sample stained according to a second staining technique.

10. The process according to claim 1 , wherein the first digital image is obtained from an optical image, an image in fluorescence or an image in polarization.

11. The process according to claim 1 , wherein X 2 is obtained from the further digital image and at least one X i with i>2 is obtained from an other digital image.

12. The process according to claim 1 , wherein X 1 is obtained from a combination of the first digital image with an other digital image and at least one X i for 2≤i≤n is obtained from a combination of the further digital image with the other digital image.

13. A data processing device comprising means for carrying out the process according to claim 1 .

14. A non-transitory computer-readable medium comprising instructions which, when executed by a data processing device, cause the data processing device to carry out the process according to claim 1 .

Assignees

• Univ Vilnius
• Univ Caen
• Inst Nat Sante Rech Med
• Centre Regional Francois Baclesse

Inventors

• Poulain Laurent
• Laurinavicius Arvydas
• Plancoulaine Benoît
• Elie Nicolas

Classifications

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title

• G06T2207/30024

  Cell structures in vitro; Tissue sections in vitro · CPC title

• G06F16/55

  Clustering; Classification · CPC title

• G06T7/60

  Analysis of geometric attributes · CPC title