Computer-aided detection of breast masses on mammograms: performance improvement using a dual system

摘要

We have developed a computer-aided detection (CAD) system for breast masses on mammograms. In thisstudy, our purpose was to improve the performance of our mass detection system by using a new dual system approachwhich combines a CAD system optimized with ”average” masses with another CAD system optimized with subtlemasses. The latter system is trained to provide high sensitivity in detecting subtle masses. For an unknownmammogram, the two systems are used in parallel to detect suspicious objects. A feed-forward backpropagationneural network trained to merge the scores of the two linear discriminant analysis (LDA) classifiers from the twosystems makes the final decision in differentiation of true masses from normal tissue. A data set of 86 patientscontaining 172 mammograms with biopsy-proven masses was partitioned into a training set and an independent test set.This data set is referred to as the average data set. A second data set of 214 prior mammograms was used for trainingthe second CAD system for detection of subtle masses. When the single CAD system trained on the average data setwas applied to the test set, the Az for false positive (FP) classification was 0.81 and the FP rates were 2.1, 1.5 and 1.3FPs/image at the case-based sensitivities of 95%, 90% and 85%, respectively. With the dual CAD system, the Az was0.85 and the FP rates were improved to 1.7, 1.2 and 0.8 FPs/image at the same case-based sensitivities. Our resultsindicate that the dual CAD system can improve the performance of mass detection on mammograms.