A computerized automatic classification scheme for mammograms based on the assessment of fibroglandular breast tissue density

摘要

It is very important to assess fibroglandular breast tissue to define the degree of risk of the lesions being obscured by normal breast tissue. We developed an automated classification method for mammograms, in which the mammograms were divided into three regions by both the variance histogram analysis and discriminant analysis, and were classified. into four categories based on the rate of each of three regions. The classification results by the normal images showed the high agreement rate between physicians and computer. As a result of malignant images' classification in the present study, the influence of existence of mass regions to classification results is dependent on not only mass sizes but also these positions. Because the rate of different classifications of right and left images in malignant database is larger than that in normal one, it may he possible to apply this scheme for potential indication of the detection of mass lesions.