There are four areas in which AI must excel to become clinically viable in women’s medical imaging—particularly mammography: performance, time, workflow and cost, according to an opinion article published in the American Journal of Roentgenology.
“AI holds tremendous potential for transforming the practice of radiology, but certain metrics are needed to objectively quantify its impact,” wrote authors Ray C. Mayo, MD, and Jessica W.T. Leung, MD, of the University of Texas MD Anderson Cancer Center in Houston. “As patients, physicians, hospitals and insurance companies look for value, AI must earn a role in medical imaging.”
Mayo and Leung noted the most important thing AI must do in women’s imaging and mammography is improve imaging performance—which includes decreasing false positives and unnecessary workups and biopsies.
“If AI earns a place in breast imaging, its best opportunity is to decrease false-positive flags compared with the number associated with currently available (computer-aided detection) programs,” the authors wrote.
When used in women’s imaging and mammography, AI should not upturn interpretation time. AI can decrease interpretation time by ensuring there are fewer “flags” to review, so radiologists can use their time for studies where quick and timely intervention is necessary.
“If even a small percentage of cases can be identified as negative with 100 percent certainty, a substantial amount of valuable physician time can be directed at more complicated cases and other activities requiring human intervention,” Mayo and Leung wrote.
AI must provide seamless workflow integration—and the AI effort should begin before image acquisition by looking through medical records to identify patients in need of specific imaging. Then the ordering process and scheduling can be spearheaded automatically by the ordering provider. Additionally, the authors wrote, the radiologist’s review of the AI image analysis should be integrated into the post-acquisition process.
“A potential application of AI to improve both work flow and patient care is the concept of rereading, whereby the algorithm begins to review archived recent prior examinations not subjected to initial AI CAD evaluation,” the researchers wrote. “Any cases flagged would be automatically moved into a queue to be reviewed by a radiologist.”
Furthermore, AI should not be utilized on separate workstations or monitors, and AI algorithms should be compatible and fully integrated with multiple manufacturers’ hardware and software systems.
Though AI is a useful tool in mammography, “the cost to provide AI interpretive assistance must not be so high that it tilts the value equation against its use,” the researchers wrote. Still, it should be noted that false-positive mammograms cost the healthcare system $4 billion annually. In turn, AI could decrease the overall cost of breast imaging.