Image diagnosis assisting apparatus, image diagnosis assisting system and image diagnosis assisting method

1. An image diagnosis assisting apparatus, comprising:

a processor configured to execute various programs for performing image processing on a target image; and

a memory configured to store a result of the image processing, wherein the processor executes:

processing of inputting an image of tissues or cells;

processing of extracting feature amounts of tissues or cells in the target image;

feature estimation processing of estimating feature amounts of an image dyed by another dyeing method different from the dyeing method of the target image by machine learning from image data of tissues or cells in the target image; and

determination processing of determining presence or absence of a lesion and lesion probability for each of the target images by using a plurality of feature amounts including at least the feature amounts of tissues or cells in the target image and the feature amounts of an image dyed by another dyeing method.

2. The image diagnosis assisting apparatus according to claim 1 , wherein

in the feature estimation processing, the processor estimates the feature amounts of the image dyed by another dyeing method different from the dyeing method of the target image by machine learning from the image data of the tissues or cells in the target image through estimation based on the target image.

3. The image diagnosis assisting apparatus according to claim 1 , wherein

in the determination processing, the processor determines the presence or absence of the lesion and the lesion probability by using a discriminator configured to calculate from the feature amounts of the tissues or cells in the target image, the feature amounts of the image dyed by another dyeing method different from the dyeing method of the target image by machine learning.

4. The image diagnosis assisting apparatus according to claim 1 , wherein

the processor displays a plurality of determination results depending on magnifications to determine the lesion probability based on the results in the respective magnifications.

5. An image diagnosis assisting apparatus, comprising:

a processor configured to execute various programs for performing image processing on a target image; and

a memory configured to store a result of the image processing, wherein the processor executes

processing of inputting an image of tissues or cells,

processing of extracting feature amounts of tissues or cells in the target image,

processing of generating, from the target image, an image dyed by another dyeing method different from the dyeing method of the target image by machine learning,

feature estimation processing of estimating feature amounts of tissues or cells in the generated image, and

determination processing of determining presence or absence of a lesion and lesion probability for each of the target images by using a plurality of feature amounts including at least the feature amounts of tissues or cells in the target image and the feature amounts of an image dyed by another dyeing method.

6. The image diagnosis assisting apparatus according to claim 5 , wherein

the processor displays a plurality of determination results depending on magnifications to determine the lesion probability based on the results in the respective magnifications.

7. An image diagnosis assisting apparatus, comprising:

a processor configured to execute various programs for performing image processing on a target image; and

a memory configured to store a result of the image processing, wherein the processor executes

processing of inputting an image of tissues or cells,

processing of extracting feature amounts of tissues or cells in the target image,

processing of generating, from the target image, an image dyed by another dyeing method different from the dyeing method of the target image by machine learning,

feature estimation processing of estimating feature amounts of tissues or cells in the generated image, and

determination processing of determining presence or absence of a lesion and lesion probability for each of the target images by using the feature amounts extracted by the processing of extracting and the feature amounts estimated by the feature estimation processing, wherein

in the determination processing, the processor determines the presence or absence of a lesion and the lesion probability by using a discriminator configured to calculate, from the feature amounts of the tissues or cells in the image, the feature amounts of the image dyed by another dyeing method different from the dyeing method of the target image by machine learning.

8. An image diagnosis assisting method for classifying desired tissues or cells in a target image, comprising:

inputting an image of tissues or cells by a processor configured to execute various programs for performing image processing on the target image;

extracting, by the processor, feature amounts of tissues or cells in the target image;

feature estimating, by the processor, feature amounts of an image dyed by another dyeing method different from the dyeing method of the target image by machine learning from the image data of tissues or cells in the target image; and

determining, by the processor, presence or absence of a lesion and lesion probability for each of the target images by using a plurality of feature amounts including at least the feature amounts of tissues or cells in the target image and the feature amounts of an image dyed by another dyeing method.

9. The image diagnosis assisting method according to claim 8 , wherein

in the feature estimating, the processor estimates the feature amounts of the image dyed by another dyeing method different from the dyeing method of the target image by machine learning from the image data of the tissues or cells in the target image through estimation based on the target image.

10. The image diagnosis assisting method according to claim 8 , wherein

the processor determines the presence or absence of the lesion and the lesion probability by using a discriminator configured to calculate from the feature amounts of the tissues or cells in the target image, the feature amounts of the image dyed by another dyeing method different from the dyeing method of the target image by machine learning.

11. The image diagnosis assisting method according to claim 8 , wherein

the processor displays a plurality of determination results depending on magnifications to determine the lesion probability based on the results in the respective magnifications.

12. An image diagnosis assisting method for classifying a desired tissues or cells in a target image, comprising:

inputting an image of tissues or cells by a processor configured to execute various programs for performing image processing on the target image;

extracting, by the processor, feature amounts of tissues or cells in the target image;

generating, by the processor, from the target image, an image dyed by another dyeing method different from the dyeing method of the target image by machine learning;

feature estimating, by the processor, feature amounts of tissues or cells in the generated image; and

determining, by the processor, presence or absence of a lesion and lesion probability for each of the target images by using a plurality of the feature amounts including at least the feature amounts of tissues or cells in the target image and the feature amounts of an image dyed by another dyeing method.

13. An image diagnosis assisting method for classifying desired tissues or cells in a target image, comprising:

inputting an image of tissues or cells by a processor configured to execute various programs for performing image processing on the target image;

extracting, by the processor, feature amounts of tissues or cells in the target image;

generating, by the processor, from the target image, an image dyed by another dyeing method different from the dyeing method of the target image by machine learning;

feature estimating, by the processor, feature amounts of tissues or cells in the generated image; and

determining, by the processor, presence or absence of a lesion and lesion probability for each of the target images by using a plurality of the feature amounts, wherein

in the determining, the processor determines the presence or absence of a lesion and the lesion probability by using a discriminator configured to calculate, from the feature amounts of the tissues or cells in the target image, the feature amounts of the image dyed by another dyeing method different from the dyeing method of the target image by machine learning.

14. A remote diagnosis assisting system, comprising:

a server including an image diagnosis assisting apparatus,

the image diagnosis assisting apparatus including

a processor configured to execute various programs for performing image processing on a target image, and

a memory configured to store a result of the image processing,

the processor executing

processing of inputting an image of tissues or cells,

processing of extracting feature amounts of tissues or cells in the target image,

feature estimation processing of estimating feature amounts of an image dyed by another dyeing method different from the dyeing method of the target image by machine learning from the image data of tissues or cells in the target image, or of generating, from the target image, an image dyed by another dyeing method different from the dyeing method of the target image by machine learning and extracting feature amounts of tissues or cells in the generated image, and

determination processing of determining presence or absence of a lesion and lesion probability for each of the target images by setting a classification based on the result and using the feature amounts extracted by the processing of extracting and the feature amounts estimated by the feature estimation processing; and

an image acquiring apparatus including an imaging apparatus configured to capture image data, wherein

the image acquiring apparatus sends the image data to the server,

the server processes, by the image diagnosis assisting apparatus, the image data that the server has received, and stores, in the memory, the image of the tissues or cells on which the determination has been made and a result of the determination and sends the image of the tissues or cells on which the determination has been made and the result of the determination to the image acquiring apparatus, and

the image acquiring apparatus displays, on a display apparatus, the image of the tissues or cells on which the determination has been made and the result of the determination that the image acquiring apparatus has received.

15. An online contract service providing system, comprising:

a server including an image diagnosis assisting apparatus, the image diagnosis assisting apparatus including

a processor configured to execute various programs for performing image processing on a target image, and

a memory configured to store a result of the image processing,

the processor executing

processing of inputting an image of tissues or cells,

processing of extracting feature amounts of tissues or cells in the target image,

feature estimation processing of estimating feature amounts of an image dyed by another dyeing method different from the dyeing method of the target image by machine learning from the image data of tissues or cells in the target image, or of generating, from the target image, an image dyed by another dyeing method different from the dyeing method of the target image by machine learning and extracting feature amounts of tissues or cells in the generated image, and

determination processing of determining presence or absence of a lesion and lesion probability for each of the target images by setting a classification based on the result and using the feature amounts extracted by the processing of extracting and the feature amounts estimated by the feature estimation processing; and

an image acquiring apparatus including

an imaging apparatus configured to capture the image data, and

the image diagnosis assisting apparatus, wherein

the image acquiring apparatus sends the image data to the server,

the server processes, by the image diagnosis assisting apparatus, the image data that the server has received, and stores, in the memory, the image of the tissues or cells on which the determination has been made and a discriminator and sends the image of the tissues or cells on which the determination has been made and the discriminator to the image acquiring apparatus,

the image acquiring apparatus stores the image of the tissues or cells on which the determination has been made and the discriminator that the image acquiring apparatus has received, and

the image diagnosis assisting apparatus in the image acquiring apparatus makes a determination on an image of another tissues or cells by using the discriminator, and displays a result of the determination on a display apparatus.