AI's use continues to rise in breast imaging, with new tools being deployed for breast cancer detection and supplemental imaging. Researchers continue to highlight how these AI tools can assist radiologists in interpreting mammograms and lessening workflow burdens.
But survey results published in August in Clinical Imaging suggest that practice managers should temper their expectations when implementing AI into their breast imaging workflows.
Azadeh Elmi, MD, from the University of California, San Diego Health joins the show to share her team's results and what they could mean for breast imagers considering AI tools for their departments and practices.
Azadeh Elmi, MD, discusses the need for better collaboration between breast radiologists and AI developers to create tools that will lead to meaningful downstream impact for patient care.
Elmi et al found that AI users reported “more modest” perceived clinical impact than anticipated. Few survey respondents reported downstream improvements in callback rate, biopsy rate, or burnout.
Elmi also weighs in on whether the difference in expectation versus actual impact is more on the radiologists or the vendors.
Watch the full episode below.
