A team of multi-institutional researchers, including UC San Francisco's Department of Radiology and Biomedical Imaging and Center for Intelligent Imaging (ci2), investigated mammogram data using artificial intelligence.
The researchers, led by Amie Lee, MD of the University of California San Francisco and ci2 member, present their findings in "Clinical Implementation of AI in Screening Mammography: The Essential Role of Prospective Evaluation," published in RSNA Radiology. Her co-author is Sarah M, Friedwald, MD of Northwestern Feinberg School of Medicine.
The authors compared the performance metrics and workload of 60,751 breast imaging scans before and after applying artificial intelligence systems. Mammography is currently the only procedure used for breast cancer screening, and it can be time-consuming.
"Other shortcomings of mammography include the inherently subjective nature of interpretation, the inevitability of human error, and the highly variable levels of performance and experience across radiologists," the authors write. "The utility of AI as a decision support tool (eg. aiding the radiologist in cancer detection or worklist triaging) and as a second reader are areas of considerable interest."
Read more about research and news at UCSF ci2.
