Compared to standard double reading from two radiologists, artificial intelligence (AI)–supported mammography screening detected one additional breast cancer per 1,000 patients, researchers reported in Lancet Oncology. AI mammography also cut radiologists’ screening workload nearly in half.
For the randomized, controlled, population-based trial, the researchers assigned 80,033 patients to receive either AI-supported mammography screening (n = 40,003) or double reading without AI (n = 40,030). They found that AI and double screening detected 6.1 and 5.1 cancers per 1,000 patients, with a 1.5% false-positive rate in both groups. They also found that AI mammography reduced the number of radiologist screen readings from 83,231 to 46,345 or a total of 44%.
“The greatest potential of AI right now is that it could allow radiologists to be less burdened by the excessive amount of reading,” the researchers said. “While our AI-supported screening system requires at least one radiologist in charge of detection, it could potentially do away with the need for double reading of the majority of mammograms, easing the pressure on workloads and enabling radiologists to focus on more advanced diagnostics while shortening waiting times for patients.”
Learn more about breast cancer screening in ONS’s Breast Cancer: Prevention, Detection, and Pathophysiology course and get additional resources for breast cancer care in the Breast Cancer Learning Library.