Noninferiority Randomized Trial of AI-Augmented Mammography Reading in Breast Cancer Screening
“Our study is the first randomized controlled trial investigating the use of AI in breast cancer screening and the largest to date looking at AI use in cancer screening in general,” said study author Kristina Lång, PhD.
To evaluate the impact of AI-supported mammography readings on the rate of interval cancers in breast cancer screening, specifically comparing AI-supported readings to standard double readings.
Key Findings:
Interval cancer rate was 1.55 per 1,000 in the AI group vs. 1.76 in the control group, with statistical significance noted.
AI-supported readings detected 29% more cancers without increasing false positives.
Sensitivity was higher in the AI group (80.5%) compared to the control (73.8%), with P-value included.
Fewer interval cancers with unfavorable characteristics were diagnosed in the AI group.
Interpretation:
AI-supported mammography screening improves early detection of clinically relevant breast cancers, reduces the workload on radiologists, and emphasizes the necessity of human oversight.
Limitations:
The study does not support replacing radiologists with AI; human oversight is still required.
The trial's findings may not be generalizable to all populations, and potential biases should be considered.
Conclusion:
AI can enhance mammography screening efficiency and effectiveness, potentially alleviating pressure on radiologists, while underscoring the need for human involvement.
ASCO AI in Oncology
is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO®). The ideas and opinions expressed in
ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.