Method for detecting the presence of different antinuclear antibody fluorescence pattern types and apparatus for this purpose

1. A method for detecting respective potential presence of respective different cellular fluorescence pattern types on a biological cell substrate comprising human epithelioma cells,

wherein the cellular fluorescence pattern types comprise a plurality of different antinuclear antibody fluorescence pattern types,

wherein the method comprises

incubating the biological cell substrate with a liquid patient sample, which potentially comprises

primary antibodies,

a first fluorescent dye, and

secondary antibodies, which are labelled with a second fluorescent dye,

acquiring a first image which represents staining of the biological cell substrate by the first fluorescent dye and acquiring a second image which represents staining of the biological cell substrate by the second fluorescent dye,

detecting, on the basis of the first image, respective image segments which in each case represent at least one mitotic cell,

selecting, on the basis of the detected image segments, only subimages of the first image and subimages corresponding thereto of the second image, the subimages corresponding thereto of the second image having pixel coordinates identical to those of the subimages of the first image, and wherein selection of the subimages corresponding thereto of the second image is based on the pixel coordinates, and

detecting, on the basis of the subimages of the first image and of the subimages of the second image, respective actual presence of respective cellular fluorescence pattern types by a convolutional neural network.

2. The method according to claim 1 , further comprising:

detecting, on the basis of the first image, respective image segments which in each case represents at least one mitotic cell of a sufficient quality,

selecting subimages of the first image and subimages corresponding thereto of the second image on the basis of the respective image segments which in each case represents at least one mitotic cell of a sufficient quality.

3. The method according to claim 1 , further comprising:

determining, on the basis of the selected subimages of the first image and of the selected subimages of the second image, respective confidence measures for the respective actual presence of the respective cellular fluorescence pattern types by the convolutional neural network.

4. The method according to claim 3 , wherein the convolutional neural network comprises:

an output layer which generates a respective feature map for the respective cellular fluorescence pattern type, and

wherein the convolutional neural network determines the respective confidence measure on the basis of the respective feature map.

5. The method according to claim 3 , further comprising:

segmenting the first image into image segments of different segment classes,

determining at least one lightness value for at least one cellular fluorescence pattern type on the basis of one or more image segments of at least one specific segment class, and

verifying the confidence measure of the at least one cellular fluorescence pattern type on the basis of the at least one lightness value of the at least one cellular fluorescence pattern type.

6. The method according to claim 5 , wherein the confidence measure is verified on the basis of the at least one lightness value and as a function of a user-predeterminable threshold value.

7. The method according to claim 3 , further comprising:

for a respective subimage tuple which comprises the subimage of the first image and the subimage of the second image, determining respective subimage confidence measures for respective actual subimage presence of respective cellular fluorescence pattern types by the convolutional neural network, and

determining the respective confidence measures for the respective actual presence of the respective cellular fluorescence pattern types on the basis of the respective subimage confidence measures.

8. An apparatus for detecting respective potential presence of respective different cellular fluorescence pattern types on a biological cell substrate comprising human epithelioma cells by digital image processing, the apparatus comprising:

a holder for the biological cell substrate, which has been incubated with a liquid patient sample and comprises,

primary antibodies,

a first fluorescent dye, and

secondary antibodies which are labelled with a second fluorescent dye,

at least one image acquisition unit for acquiring a first image which represents staining of the biological cell substrate by the first fluorescent dye and for acquiring a second image which represents staining of the biological cell substrate by the second fluorescent dye, and

at least one computing unit, which is configured

to detect, on the basis of the first image, respective image segments which in each case represents at least one mitotic cell,

to select, on the basis of the respective image segments, only subimages of the first image and subimages corresponding thereto of the second image, the subimages corresponding thereto of the second image having pixel coordinates identical to those of the subimages of the first image, and wherein selection of the subimages corresponding thereto of the second image is based on the pixel coordinates, and

to detect, on the basis of the subimages of the first image and of the subimages of the second image, respective actual presence of respective cellular fluorescence pattern types by a convolutional neural network.

9. A method for detecting respective potential presence of respective different cellular fluorescence pattern types on a biological cell substrate comprising human epithelioma cells by digital image processing, the method comprising:

acquiring a first image which represents staining of a biological cell substrate by a first fluorescent dye,

acquiring a second image which represents staining of the biological cell substrate by a second fluorescent dye,

detecting, on the basis of the first image, respective image segments which in each case represent at least one mitotic cell,

selecting, on the basis of the respective image segments, only subimages of the first image and subimages corresponding thereto of the second image, the subimages corresponding thereto of the second image having pixel coordinates identical to those of the subimages of the first image, and wherein selection of the subimages corresponding thereto of the second image is based on the pixel coordinates, and

detecting, on the basis of the subimages of the first image and of the subimages of the second image, respective actual presence of respective cellular fluorescence pattern types by a convolutional neural network.

10. A method for digital image processing, comprising:

receiving a first image which represents staining of a biological cell substrate by a first fluorescent dye, wherein the biological cell substrate comprises human epithelioma cells,

receiving a second image which represents staining of the biological cell substrate by a second fluorescent dye,

detecting, on the basis of the first image, respective image segments which in each case represents at least one mitotic cell,

selecting, on the basis of the detected image segments, only subimages of the first image and subimages corresponding thereto of the second image, the subimages corresponding thereto of the second image having pixel coordinates identical to those of the subimages of the first image, and wherein selection of the subimages corresponding thereto of the second image is based on the pixel coordinates, and

detecting, on the basis of the subimages of the first image and of the subimages of the second image, respective actual presence of respective cellular fluorescence pattern types by a convolutional neural network.

11. A computing unit which, in a course of digital image

processing, is configured to:

receive a first image, which represents staining of a biological cell substrate by a first fluorescent dye, wherein the biological cell substrate comprises human epithelioma cells,

receive a second image, which represents staining of the biological cell substrate by a second fluorescent dye,

detect, on the basis of the first image, respective image segments which in each case represents at least one mitotic cell,

select, on the basis of the respective image segments, only subimages of the first image and subimages corresponding thereto of the second image, the subimages corresponding thereto of the second image having pixel coordinates identical to those of the subimages of the first image, and wherein selection of the subimages corresponding thereto of the second image is based on the pixel coordinates, and

detect, on the basis of the subimages of the first image and of the subimages of the second image, respective actual presence of respective cellular fluorescence pattern types by a convolutional neural network.

12. A data network apparatus, comprising:

at least one data interface for receiving the first image which represents staining of the biological cell substrate by the first fluorescent dye, wherein the biological cell substrate comprises human epithelioma cells, and for receiving the second image which represents staining of the biological cell substrate by the second fluorescent dye, and

the computing unit according to claim 11 .

13. A non-transitory machine-readable medium having stored thereon machine-readable instructions executable to cause a machine to perform operations comprising the method for digital image processing according to claim 10 .

14. The method according to claim 1 , wherein selected subimages of the first image and subimages corresponding thereto of the second image each contain the at least one mitotic cell.

15. The method according to claim 1 , wherein the first image and the second image have identical spatial orientation.