Computer-aided detection systems show high potential for detecting inconspicuous colonic polyps but provide little help in locating nonpolypoid colorectal lesions, according to research presented Wednesday at the RSNA meeting.
Computer-aided detection systems show high potential for detecting inconspicuous colonic polyps but provide little help in locating nonpolypoid colorectal lesions, according to research presented Wednesday at the RSNA meeting.
Results of a study of 411 patients with three of the most conspicuous types of polyps -- near folds, on folds, and on walls -- showed high sensitivity for all three types, regardless of the polyp's conspicuity.
However, another study in the session, comparing the sensitivity of CT colonoscopies read by human readers with computer-aided diagnoses for detecting nonpolypoid (flat) colorectal lesions, showed that human readers had a 67% detection rate, compared to 56% for CAD.
The study noted, however, that using CAD software may help lower human perceptive errors and reduce reporting time.
AI-Initiated Recalls After Screening Mammography Demonstrate Higher PPV for Breast Cancer
March 18th 2025While recalls initiated by one of two reviewing radiologists after screening mammography were nearly 10 percent higher than recalls initiated by an AI software, the AI-initiated recalls had an 85 percent higher positive predictive value for breast cancer, according to a new study.