PB.23: Effect of detector type on cancer detection in digital mammography

Abstract

This work measured the effect that image quality associated with different detectors has on cancer detection in mammography using a novel method for changing the appearance of images. A set of 270 mammography cases (one view, both breasts) was acquired using five Hologic Selenias and two Hologic Dimensions X-ray units: 80 normal, 80 with simulated inserted subtle calcification clusters, 80 with subtle real noncalcification malignant lesions and 30 with benign lesions (biopsy proven). These 270 cases (Arm 1) were converted to appear as if they had been acquired on two other imaging systems: needle image plate computed radiography (CR) (Arm 2) and powder phosphor CR (Arm 3). Three experienced mammography readers marked the location of suspected cancers in the images and classified whether each lesion would require further investigation and the confidence in that decision. Performance was calculated as the area under curve (AUC) of the alternative free-response receiver operating characteristic curve