A Computer-aided Diagnosis System for Digital Mammograms Based on the Detection of Microcalcifications
Lixin Song · Harbin Ligong Daxue xuebao · 2006
To the question that the microcalcifications in the mammography are extremely small,various shapes,different sizes,variable distributions,and low contrast.The intensity difference between suspicious areas and their surrounding tissues can be quite slim.According to the principle that microcalcificatins are relatively high-frequency components buried in the background of low-frequency components and very high-frequency noise in the mammograms,the paper presents a method to detec the microcalcifications by morphology and wavelet transform.And according to the principle that artificial neural network can implement classification through training,the paper presents a method to identificate the types of pathological changes by probability neural network.Experiments indicate that the algorithm not only has simple operation,and achieves high true positive detection rate at the cost of low false positive.