A GENETIC ALGORITHM-BASED METHOD FOR OPTIMIZING THE PERFORMANCE OF A COMPUTER-AIDED DIAGNOSIS SCHEME FOR DETECTION OF CLUSTERED MICROCALCIFICATIONS IN MAMMOGRAMS

Computer-aided diagnosis (CAD) schemes have the potential of substanti ally increasing diagnostic accuracy in mammography by providing the ad vantages of having a second reader. Our laboratory has developed a CAD scheme for detecting clustered microcalcifications in digital mammogr ams that is being tested clinically at the University of Chicago Hospi tals. Our CAD scheme contains a large number of parameters such as fil ter weights, threshold levels, and region of interest (ROI) sizes. The choice of these parameter values determines the overall performance o f the system and thus must be carefully set. Unfortunately, when the n umber of parameters becomes large, it is very difficult to obtain the optimal performance, especially when the values of the parameters are correlated with each other. In this study, we address the problem of i dentifying the optimal overall performance by developing an automated method for the determination of the parameter values that maximize the performance of a mammographic CAD scheme. Our method utilizes a genet ic algorithm to search through the possible parameter values, and prov ides the set of parameters that minimize a cost function which measure s the performance of the scheme. Using a database of 89 digitized mamm ograms, our method demonstrated that the sensitivity of our CAD scheme can be increased from 80% to 87% at a false positive rate of 1.0 per image. We estimate the average performance of our CAD scheme on unknow n cases by performing jackknife tests; this was previously not feasibl e when the parameters of the CAD scheme were determined in a nonautoma ted manner. (C) 1998 American Association of Physicists in Medicine.