Changes in cancer detection and false-positive recall in mammography using artificial intelligence: a retrospective, multireader study.
There is increasing interest in the use of artificial intelligence (AI) to improve breast cancer detection. This study developed and validated an AI algorithm using mammography readings from five institutions in South Korea, the United States, and the United Kingdom. The AI algorithm alone showed better diagnostic performance in breast cancer detection compared to radiologists without AI assistance (area under the curve [AUC] of 0.94 vs. 0.81, p<0.0001) or radiologists with AI assistance (0.88; p<0.0001). AI improved performance of radiologists and was better at detecting mass cancers, distortion, asymmetry, or node-negative cancers compared with radiologist alone.