Getting a mammogram is vital for breast cancer screening; however, it is not 100% accurate. In some cases, it could give false negatives. To address this issue, UC Davis Health is conducting the Pragmatic Randomized Trial of Artificial Intelligence for Screening Mammography, or PRISM trial, to arrive at a solution. Their research is part of a national clinical trial to determine whether AI can help radiologists interpret screening mammograms more accurately.
Testing AI effectiveness
The PRISM has a funding of $16 million thanks to the Patient-Centered Outcomes Research Institute. Diana Miglioretti, dual principal investigator and lead of the study’s data coordinating center, said that this “large-scale randomized trial in the US” aims to evaluate whether AI-assisted interpretation would get better outcomes or not.
Talking more about how the PRISM trial results will be used, Miglioretti said that the results would help them to understand clinical practice around mammography, technology adoption, and how they are informing the patients about AI in screening. While UCLA would act as the administrative coordinating site, the PRISM trial includes other leading universities in the field, like UC San Diego, Boston Medical Center, University of Miami, University of Washington, and University of Wisconsin.
Furthermore, the PRISM trial is supported by patient advocates, clinicians, health system leaders, and policymakers. In conducting the trials, facilities were asked to carry out the usual screenings. Mammograms will be randomly assigned to be either interpreted by a radiologist or done with the help of an AI tool. However, the final outcome will ultimately be worked out by an expert.
Assistance, not replacements
The PRISM trial is underway, so it’s important to note that this is to determine how accurate AI assistance could be when determining something as life-threatening as cancer. Miglioretti emphasized that their aim was not to replace radiologists with AI but to know how accurate AI could be as a co-pilot in reading mammography. Other than cancer detection, they also plan to have focus groups and surveys to know what the patients and radiologists think of AI-assisted care.
YouTube: UC Davis to co-lead $16 million study axamining AI’s role in reading mammograms
By clicking play, you agree to YouTube's Terms of Service and Privacy Policy. Data may be shared with YouTube/Google.
Photo credit: The feature image has been taken by the National Cancer Institute.
