Most patients are open to AI assisting with mammogram interpretation, but only when a radiologist is involved.
More than 70% of patients surveyed supported the use of AI to help read mammograms, although support dropped sharply when AI was presented as a standalone tool, according to a Breast Cancer Research and Treatment study cited in a Jan. 20 news release from UT Southwestern Medical Center in Dallas. Only 6.6% of respondents favored AI acting as the sole reader, which researchers said underscores continued trust in physician oversight.
The survey included 924 patients who received mammograms at UT Southwestern’s William P. Clements Jr. University Hospital and Parkland Health, a public safety-net system also based in Dallas.
Here are five key findings from the study:
- Nearly 60% of respondents said they would rather wait hours or days for a radiologist’s interpretation than receive immediate results from AI alone. Researchers said the response highlights the importance patients place on human judgment and the clinician-patient relationship.
- Support for AI use was largely consistent across hospital settings after researchers accounted for demographic factors such as age, education, income and race. While initial results showed lower approval among patients at Parkland Health, those differences were no longer significant after adjustment.
- About 74% of participants said they would want to be informed or give consent before AI is used to help interpret their mammograms. More than 80% reported concern about at least one issue related to AI, including data privacy, bias, accuracy and potential effects on the physician-patient relationship.
- Patient preferences varied depending on how AI was positioned in the clinical workflow. While 84% said they wanted a radiologist to review abnormalities flagged by AI, only 44% supported AI reviewing findings identified by a radiologist.
- Non-Hispanic Black participants were less likely to accept AI use and more likely to express privacy concerns. Researchers emphasized the need for culturally sensitive communication as AI tools are introduced into breast cancer screening.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.