Learning system, method, and program
Provided are a learning system, a method, and a program that can provide medical institutions with motivation for creating learning data related to medical images in a case of using machine learning to support diagnosis using medical images. The learning system includes a reception unit that receives an input of first learning data from a user, a calculation unit that calculates, a contribution degree of the first learning data to learning of a discriminator for each user on the basis of at least one of a comparison result between the first learning data and second learning data used for creating the discriminator or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data, and a service setting unit that sets a service for the user on the basis of the contribution degree calculated for each user.
1. A learning system comprising:
a processor configured to
receive an input of first learning data from a user, the first learning data comprising an image and analysis result information of the image;
calculate a contribution degree of the first learning data to learning of a discriminator for each user,
wherein the discriminator is an image identification engine that is created by second learning data, the second learning data comprising another image and analysis result information of the another image,
wherein the contribution degree is calculated on the basis of at least one of a comparison result between the first learning data and the second learning data or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data; and
set a service for the user on the basis of the contribution degree calculated for each user,
wherein when the processor calculates the contribution degree on the basis of the comparison result between the first learning data and the second learning data, the processor calculates the contribution degree on the basis of a difference between a comparison data set including the second learning data and the first learning data, and on the basis of a difference between a feature vector obtained from an image included in the second learning data and a feature vector obtained from an image included in the first learning data.
2. The learning system according to claim 1 ,
wherein the processor is further configured to receive an input of the correct answer data for an output obtained by inputting the first learning data to the discriminator, and
wherein the processor calculates the contribution degree on the basis of a result of comparing the output with the correct answer data.
3. The learning system according to claim 2 ,
wherein the processor receives an input of correction of at least one of a contour or a size of a region extracted from an image included in the first learning data by the discriminator, and
wherein the processor calculates the contribution degree on the basis of an amount of the correction.
4. The learning system according to claim 1 ,
wherein the processor calculates a difference in accuracy of an image analysis result in the discriminator before and after learning using the first learning data, and calculates the contribution degree of the first learning data to learning of the discriminator for each user on the basis of the difference in accuracy.
5. The learning system according to claim 1 ,
wherein the processor receives an input of data including a medical image of a patient as the first learning data, and
wherein the processor is further configured to create and store data in which identification information capable of identifying the patient is concealed, in the first learning data.
6. A learning method performed in a learning system comprising:
receiving an input of first learning data from a user, the first learning data comprising an image and analysis result information of the image;
calculating a contribution degree of the first learning data to learning of a discriminator for each user,
wherein the discriminator is an image identification engine that is created by second learning data, the second learning data comprising another image and analysis result information of the another image,
wherein the contribution degree is calculated on the basis of at least one of a comparison result between the first learning data and the second learning data or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data; and
setting a service for the user on the basis of the contribution degree calculated for each user,
wherein when the contribution degree is calculated on the basis of the comparison result between the first learning data and the second learning data, the contribution degree is calculated on the basis of a difference between a comparison data set including the second learning data and the first learning data, and on the basis of a difference between a feature vector obtained from an image included in the second learning data and a feature vector obtained from an image included in the first learning data.
7. A non-transitory, computer-readable recording medium which records therein, computer instructions that, when executed by a computer, causes the computer to realize:
a function of receiving an input of first learning data from a user, the first learning data comprising an image and analysis result information of the image;
a function of calculating a contribution degree of the first learning data to learning of a discriminator for each user,
wherein the discriminator is an image identification engine that is created by second learning data, the second learning data comprising another image and analysis result information of the another image,
wherein the contribution degree is calculated on the basis of at least one of a comparison result between the first learning data and the second learning data or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data; and
a function of setting a service for the user on the basis of the contribution degree calculated for each user,
wherein when the contribution degree is calculated on the basis of the comparison result between the first learning data and the second learning data, the contribution degree is calculated on the basis of a difference between a comparison data set including the second learning data and the first learning data, and on the basis of a difference between a feature vector obtained from an image included in the second learning data and a feature vector obtained from an image included in the first learning data.
8. A learning system comprising:
a processor configured to
receive an input of first learning data from a user, the first learning data comprising an image and analysis result information of the image;
calculate a contribution degree of the first learning data to learning of a discriminator for each user,
wherein the discriminator is an image identification engine that is created by second learning data, the second learning data comprising another image and analysis result information of the another image,
wherein the contribution degree is calculated on the basis of at least one of a comparison result between the first learning data and the second learning data or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data; and
set a service for the user on the basis of the contribution degree calculated for each user,
wherein when the processor calculates the contribution degree on the basis of the comparison result between the first learning data and the second learning data, the processor calculates the contribution degree on the basis of a difference between average data created from the second learning data used for creating the discriminator and the first learning data.
9. A learning system comprising:
a processor configured to
receive an input of first learning data from a user, the first learning data comprising an image and analysis result information of the image;
calculate a contribution degree of the first learning data to learning of a discriminator for each user,
wherein the discriminator is an image identification engine that is created by second learning data, the second learning data comprising another image and analysis result information of the another image,
wherein the contribution degree is calculated on the basis of at least one of a comparison result between the first learning data and the second learning data or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data; and
set a service for the user on the basis of the contribution degree calculated for each user,
wherein when the processor calculates the contribution degree on the basis of the comparison result between the first learning data and the second learning data, the processor calculates the contribution degree on the basis of a difference between average data created from the second learning data used for creating the discriminator and the first learning data, and on the basis of a difference between a feature vector obtained from an average image created from an image included in the second learning data and a feature vector obtained from an image included in the first learning data.