Kind Code: A1 Abstract: A similarity determination apparatus, method, and program allow a similarity between images to be appropriately determined according to the size of a finding included in the images. A finding classification unit (22) classifies each pixel of a first medical image into at least one of a plurality of types of findings. The feature value calculation unit 23 calculates a first feature value for each finding. The weighting factor setting unit 24 sets, for each finding, a weighting factor having a different degree of weighting according to the size of each finding. The adjusting unit 25 derives an adjustment weight coefficient by adjusting the weight coefficient for each of a plurality of finding groups in which a plurality of types of findings are classified. The similarity deriving unit 26 determines, based on the adjustment weighting coefficient, the first feature amount for each finding calculated in the first medical image and the second feature amount for each finding calculated in advance in the second medical image. A similarity between the first medical image and the second medical image is derived by performing a weighting operation with the feature amount. [Selection diagram] FIG.【課題】類似度決定装置、方法およびプログラムにおいて、画像に含まれる所見のサイズに応じて適切に画像間の類似度を決定できるようにする。【解決手段】所見分類部２２が、第１の医用画像の各画素を複数種類の所見のうちの少なくとも１つの所見に分類する。特徴量算出部２３が、所見毎に第１の特徴量を算出する。重み係数設定部２４が、各所見に対して、各所見のサイズに応じて重み付けの程度が異なる重み係数を設定する。調整部２５が、複数種類の所見が分類されてなる複数の所見グループ毎に、重み係数を調整して調整重み係数を導出する。類似度導出部２６が、調整重み係数に応じて、第１の医用画像において算出された所見毎の前記第１の特徴量と、第２の医用画像において予め算出された所見毎の第２の特徴量との重み付け演算を行って、第１の医用画像と第２の医用画像との類似度を導出する。【選択図】図２
