Despite its success, mammography still has a false negative.rate of 10-30% and significant variability in reading.The use of CAD/DSP methods as applied to retrospective case analysis has been widely reported with a TP detection sensitivity of about 90% and an average FP detection rate of 2-4/image, and has proved to be useful, especially for missed interval cancers.For prospective case analysis, the performance reported dropped to about 70% sensitivity with a similar FP detection rate.However, allowance should be made for the constraints of long term prospective studies, most initiated since 1995, where subsequent algorithm improvements cannot be implemented to improve perfonrmanee during the trial.The reduction in performance for prospective studies can be partly attributed to the lack of robustness of the CAD/DSP algorithms across large image data sets containing cases of varying difficulty.