APPARATUS, METHOD, AND PROGRAM FOR TRAINING DISCRIMINATOR DISCRIMINATING DISEASE REGION, DISCRIMINATOR DISCRIMINATING DISEASE REGION, DISEASE REGION DISCRIMINATION APPARATUS, AND DISEASE REGION DISCRIMINATION PROGRAM
PROBLEM TO BE SOLVED: To allow highly accurate discrimination of a disease region by the use of limited amounts of data, even in the case of an image in which it is difficult to prepare large amounts of data indicating a correct disease region, in an apparatus, method, and program for training a discriminator discriminating a disease region, a discriminator discriminating a disease region, a disease region discrimination apparatus, and a disease region discrimination program.SOLUTION: A discriminator comprises: a common learning unit； and a plurality of learning units that are connected to an output unit of the common learning unit. The discriminator is trained such that when a medical image is input to the common learning unit using a plurality of datasets of the medical image and a correct mask of a first disease region, information indicating the first disease region is output from a first learning unit. The discriminator is trained such that when a medical image is input to the common learning unit using a plurality of datasets of the medical image and a correct mask of a second disease region having at least one of medical and anatomical cause-and-effect relations with the first disease, information indicating the second disease region is output from a second learning unit.SELECTED DRAWING: Figure 2COPYRIGHT: (C)2020,JPO&INPIT【課題】疾患領域を判別する判別器の学習装置、方法及びプログラム、疾患領域を判別する判別器、並びに疾患領域判別装置及びプログラムにおいて、疾患の正解領域を示すデータを多く用意することが困難な画像であっても、限られたデータを用いて疾患領域を精度よく判別する。【解決手段】判別器は、共通学習部と、共通学習部の出力部に接続された複数の学習部と、を備える。医用画像及び第１の疾患領域の正解マスクのデータセットを複数用いて、医用画像を共通学習部へ入力した場合に、第１の学習部から第１の疾患領域を示す情報が出力されるように学習させる。医用画像及び第１の疾患と医学的及び解剖学的の少なくとも一方に因果のある第２の疾患領域の正解マスクのデータセットを複数用いて、医用画像を共通学習部へ入力した場合に、第２の学習部から第２の疾患領域を示す情報が出力されるように学習させる。【選択図】図２
