Machine learning model creation support apparatus, method of operating machine learning model creation support apparatus, and program for operating machine learning model creation support apparatus
A machine learning model creation support apparatus including a processor, in which the processor acquires a plurality of pieces of annotation information generated by a plurality of annotators assigning a plurality of labels according to a plurality of classes to a region of the same medical image, derives, for each of the classes, commonality data indicating commonality in how the labels are assigned by the plurality of annotators for the plurality of pieces of annotation information, and generates confirmed annotation information used as correct answer data of a machine learning model based on the commonality data and a preset confirmation condition.
1 . A machine learning model creation support apparatus comprising a processor,
wherein the processor
acquires a plurality of pieces of annotation information generated by a plurality of annotators assigning a plurality of labels according to a plurality of classes to a region of the same medical image,
derives, for each of the classes, commonality data indicating commonality of the labels are assigned to the region of the same medical image by the plurality of annotators for the plurality of pieces of annotation information, and
generates confirmed annotation information used as correct answer data of a machine learning model based on the commonality data and a preset confirmation condition.
2 . The machine learning model creation support apparatus according to claim 1 , wherein the annotation information is information in which the labels of the classes different from each other are assigned to the same region.
3 . The machine learning model creation support apparatus according to claim 1 , wherein:
the commonality data is a numerical value related to the number of persons of the annotators who have assigned the labels for each of the plurality of classes, and
the confirmation condition is that the assigned labels are adopted only in a case where the numerical value is more than or equal to a threshold value.
4 . The machine learning model creation support apparatus according to claim 1 , wherein display condition data for displaying the medical image under the same display condition in a case where the plurality of annotators view the medical image is attached to the medical image.
5 . The machine learning model creation support apparatus according to claim 1 , wherein the processor
detects a human body region in which a human body appears in the medical image, and
derives the commonality data only for the detected human body region.
6 . The machine learning model creation support apparatus according to claim 1 , wherein the processor performs weighting according to an attribute of the annotators in a case of deriving the commonality data.
7 . The machine learning model creation support apparatus according to claim 1 , wherein the confirmation condition is different between a central portion and a peripheral portion of the region of the classes, and a difficulty level of satisfying the condition is higher in the peripheral portion than in the central portion.
8 . The machine learning model creation support apparatus according to claim 1 , wherein the processor sets a numerical value representing reliability of the labels at a peripheral portion to be lower than a numerical value representing reliability of the labels at a central portion of the region of the classes in the confirmed annotation information.
9 . The machine learning model creation support apparatus according to claim 1 , wherein the processor transmits the annotation information and the confirmed annotation information to annotator terminals used by the annotators.
10 . A method of operating a machine learning model creation support apparatus, the method comprising:
acquiring a plurality of pieces of annotation information generated by a plurality of annotators assigning a plurality of labels according to a plurality of classes to a region of the same medical image;
deriving, for each of the classes, commonality data indicating commonality of the labels are assigned to the region of the same medical image by the plurality of annotators for the plurality of pieces of annotation information; and
generating confirmed annotation information used as correct answer data of a machine learning model based on the commonality data and a preset confirmation condition.
11 . A non-transitory computer-readable storage medium storing a program for operating a machine learning model creation support apparatus, the program causing a computer to execute processing of:
acquiring a plurality of pieces of annotation information generated by a plurality of annotators assigning a plurality of labels according to a plurality of classes to a region of the same medical image;
deriving, for each of the classes, commonality data indicating commonality of the labels are assigned to the region of the same medical image by the plurality of annotators for the plurality of pieces of annotation information; and
generating confirmed annotation information used as correct answer data of a machine learning model based on the commonality data and a preset confirmation condition.