Avenda Health, an AI healthcare company, claims its software can more accurately assess prostate cancer than human doctors. A study published last month involved ten physicians who evaluated 50 prostate cancer cases. Avenda’s Unfold AI software achieved an accuracy rate of 84.7%, compared to the manual detection rates of the doctors, which ranged from 67.2% to 75.9%.
This research, conducted in collaboration with UCLA Health and featured in the Journal of Urology, demonstrated that AI significantly enhances the precision of cancer measurement. When utilizing AI for cancer contouring, the predictions of tumor size were proven to be 45 times more accurate and consistent than when made without AI assistance.
Shyam Natarajan, an assistant adjunct professor of urology, surgery, and bioengineering at UCLA and the senior author of the study, noted that AI assistance improved both accuracy and consistency among doctors, resulting in a higher level of agreement among them.
While doctors typically rely on MRI scans to evaluate tumor size, some tumors are not visible on MRIs, according to Dr. Wayne Brisbane, an assistant professor of urology at UCLA’s David Geffen School of Medicine. In these situations, AI technologies provide critical support where imaging fails.
Dr. Brisbane emphasized that incorporating AI into cancer treatment could pave the way for more personalized and effective care, allowing for therapies better suited to individual patient needs and more successful at combating the disease. He stated that AI possesses capabilities that can “go beyond human ability.”
Avenda Health’s CEO, Dr. Natarajan, expressed his excitement over the validation of such innovations through research and recognition by the American Medical Association.
According to the American Cancer Society, approximately 1 in 8 men in the U.S. will receive a prostate cancer diagnosis in their lifetime, and 1 in 44 will succumb to the disease. This year, an estimated 299,010 new cases of prostate cancer are expected in the U.S., with 35,250 fatalities anticipated from the disease.