Superior breast cancer detection rate seen compared with standard double-reading, without negatively affecting recall rate
By Elana Gotkine HealthDay Reporter
TUESDAY, Jan. 14, 2025 (HealthDay News) — Artificial intelligence (AI)-supported double-reading mammography is associated with a higher breast cancer detection rate, without affecting the recall rate, compared with standard double reading, according to a study published online Jan. 7 in Nature Medicine.
Nora Eisemann, Ph.D., from the Institute for Social Medicine and Epidemiology at the University of Lübeck in Germany, and colleagues compared the performance of AI-supported double reading to standard reading (without AI) among women aged 50 to 69 years undergoing organized mammography screening at 12 sites in Germany. Radiologists voluntarily chose whether to use the AI system or not. A total of 463,094 women were screened (260,739 with AI support) by 119 radiologists from July 2021 to February 2023.
The researchers found that the breast cancer detection rate for radiologists in the AI-supported screening group was 6.7 per 1,000, which was 17.6 percent higher than the rate of 5.7 per 1,000 achieved in the control group. In the AI group, the recall rate was 37.4 per 1,000, which was lower than and noninferior to that seen in the control group (38.3 per 1,000). The positive predictive value of recall was 17.9 and 14.9 percent in the AI and control groups, respectively; the positive predictive value of biopsy was 64.5 and 59.2 percent, respectively.
“Our findings substantially add to the growing body of evidence suggesting that AI-supported mammography screening is feasible and safe and can reduce workload,” the authors write.
Several authors disclosed ties to medical technology companies, including Vara, which funded the study.
Copyright © 2025 HealthDay. All rights reserved.
