Image analysis method, apparatus, program, and learned deep learning algorithm

1. An image analysis method for analyzing an image of a tissue or a cell to be analyzed using a deep learning algorithm of a neural network structure, the method comprising:

generating analysis data from an analysis target image that include the tissue or the cell to be analyzed;

inputting the analysis data to the deep learning algorithm, and

generating data indicating a region of a cell nucleus in the analysis target image by the deep learning algorithm,

wherein training data used for learning of the deep learning algorithm are generated based on a bright field image of a tissue specimen or a sample containing a cell captured under a bright field microscope and a fluorescence image of a cell nucleus in the same tissue specimen or the same sample prepared by applying fluorescent nuclear stain to the same tissue specimen or the same sample for fluorescence observation by a fluorescence microscope, wherein the fluorescence image is converted to binary data indicating which part of the fluorescence image is the cell nucleus region, and

the analysis target image is a bright field image.

2. The image analysis method according to claim 1 , wherein

the analysis target image is an image of a tissue diagnostic sample, and the analysis target image includes a hue comprised of one primary color or a hue obtained by combining two or more primary colors.

3. The image analysis method according to claim 1 , wherein

the analysis target image is an image of a cell diagnostic sample, and the analysis target image includes a hue comprised of one primary color or a hue obtained by combining two or more primary colors.

4. The image analysis method according to according to claim 1 , wherein

the deep learning algorithm determines whether an arbitrary position in the analysis target image is a region of a cell nucleus.

5. The image analysis method according to claim 1 , wherein

analysis data are generated for each region of the predetermined number of pixels including peripheral pixels circumscribing a predetermined pixel; and

the deep learning algorithm generates a label indicating whether the predetermined pixel is one of the region of the cell nucleus.

6. The image analysis method according to claim 1 , wherein

a number of nodes of the input layer of the neural network corresponds to a product of the predetermined number of pixels of the analysis data and a number of combined primary colors.

7. The image analysis method according to claim 2 , wherein

the tissue diagnostic sample is a stained sample, and the analysis target image is an image obtained by imaging the stained sample under a bright field microscope.

8. The image analysis method according to claim 2 , wherein

the cell diagnostic sample is a stained sample, and the analysis target image is an image obtained by imaging the stained sample under a bright field microscope.

9. The image analysis method according to claim 7 , wherein

a stain for bright-field observation comprises hematoxylin-eosin.

10. The image analysis method of claim 8 , wherein

a stain for bright field observation is Papanicolaou.

11. The image analysis method according to claim 1 , wherein

the deep learning algorithm classifies the analysis data into classes indicating a region of a cell nucleus contained in the analysis target image.

12. The image analysis method according to claim 1 , wherein

an output layer of the neural network comprises a node having a softmax function as an activation function.

13. The image analysis method according to claim 1 , further comprising

outputting a cell nucleus region weighted image that the region of the cell nucleus is superimposed on the analysis target image.

14. An image analysis apparatus for analyzing an image of a tissue or a cell using a deep learning algorithm of a neural network structure, the apparatus comprising a system configured to:

generating analysis data from an analysis target image that include the tissue or the cell to be analyzed;

inputting the analysis data to the deep learning algorithm, and

generating data indicating a region of a cell nucleus in the analysis target image by the deep learning algorithm,

wherein training data used for learning of the deep learning algorithm are generated based on a bright field image of a tissue specimen or a sample containing a cell captured under a bright field microscope and a fluorescence image of a cell nucleus in the same tissue specimen or the same sample prepared by applying fluorescent nuclear stain to the same tissue specimen or the same sample for fluorescence observation by a fluorescence microscope, wherein the fluorescence image is converted to binary data indicating which part of the fluorescence image is the cell nucleus region, and

the analysis target image is a bright field image.