The purpose of this study is to develop an automated-detection algorithm for clustered microcalcifications on digital mammograms. The proposed system consists of the combinations of relatively simple algorithms for faster processing. The processings, such as overall background-trend correction in breast region, enhancement of high-frequency components, and contraction-expansion processing, are included. In a study of 32 clinical mammograms including 17 normal and 15 abnormal cases, our scheme achieves a 14/15(93.3%) true positive film classification (i. e. those containing clusters) with false positive clusters detected in 2/17 (11.8%) films. These results indicate that the automated method has the potential to aid physicians in screening mammograms for clustered microcalcifications.
View full abstract