To eliminate false positive candidates from abnormal shadow candidates to improve a diagnostic efficiency.

A mamma area is extracted from image data obtained from the right and left mammae of a patient, the the right and left mammae are positioned (step S2), and the the right and left mamma areas are divided into 40 to 1,000 areas (step S3), respectively. Then, the abnormal shadow candidates are detected in the respective areas (step S4), and the feature amounts of microcalcification candidates are compared in the respective corresponding areas of the right and left image data (step S5). The microcalcification candidate that has a difference between the respective featured values smaller than a preset threshold value in the comparison of the left and right image data is eliminated as a false positive candidate or a benign candidate (step S7). The microcalcification candidate that has a difference between the respective featured value equal to or more than the preset threshold value is left as a true positive candidate or malignant candidate and decided as an abnormal shadow candidate area (step S8).

COPYRIGHT: (C)2005,JPO&NCIPI