Evaluation of an automated wavelet-based system dedicated to the detectionof clustered microcalcifications in digital mammograms

Mammographic screening programs are delivering reductions in breast cancer mortality. However, breast cancer screening will be cost effective and will provide a real profit only when both high sensitivity and specificity leve ls are reached. To date, due to human or technical factors, a significant n umber of breast cancers are still missed or misinterpreted on the mammogram s. Computer methodologies, developed to assist radiologists, could represen t further amelioration by increasing diagnostic accuracy in the screening p rograms. We have tested a computerized scheme to detect clustered microcalc ifications in digital mammograms, employing 360 mammograms that were random ly selected from the mammographic screening program, currently undergoing a t the Galicia Community (Spain). After the digitization process, the breast border was initially determined. A wavelet-based algorithm was employed to detect the clusters of microcalcifications. The performance of the automat ed system over the test set was evaluated employing Free-response Receiver Operating Characteristic (FROC) methodology. The sensitivity achieved was 7 4% at a false positive detection rate of 1.83. The corresponding area under the Alternative FROC (AFROC) curve was A(1)=0.667+/-0.09.