Potential contribution of computer-aided detection to the sensitivity of screening mammography

PURPOSE: To determine the false-positive rate in screening mammography, the capability of computer-aided detection (CAD) to identify these missed lesi ons, and whether or not CAD increases the radiologists' recall rate. MATERIALS AND METHODS: All available screening mammograms that led to the d etection of biopsy-proved cancer (n = 1,083) and the most recent correspond ing prior mammograms (n = 427) were collected from 13 facilities. Panels of radiologists evaluated the retrospectively visible prior mammograms by mea ns of blinded review. All mammograms were analyzed by a CAD system that mar ks features associated with cancer. The recall rates of 14 radiologists wer e prospectively measured before and after installation of the CAD system. RESULTS: At retrospective review, 67% (286 of 427) of screening mammography detected breast cancers were visible on the prior mammograms. At independe nt, blinded review by panels of radiologists, 27% (115 of 427) were interpr eted as warranting recall on the basis of a statistical evaluation index; a nd the CAD system correctly marked 77% (89 of 115) of these cases. The orig inal attending radiologists' sensitivity was 79% (427 of [427 + 115]). Ther e was no statistically significant increase in the radiologists' recall rat e when comparing the values before (8.3%) with those after (7.6%) installat ion of the CAD system. CONCLUSION: The original attending radiologists had a false-negative rate o f 21% (115 of [427 + 115]). CAD prompting could have potentially helped red uce this false-negative rate by 77% (89 of 115) without an increase in the recall rate.