A healthcare technology company has announced that its AI software demonstrates a higher accuracy in detecting the extent of prostate cancer than human doctors.
Avenda Health published a study revealing that their Unfold AI software accurately identified prostate cancer in 84.7% of cases reviewed by doctors, while traditional methods yielded accuracy rates from 67.2% to 75.9%. The research involved ten physicians assessing 50 different prostate cancer cases and was conducted in partnership with UCLA Health, appearing in the Journal of Urology.
The study also highlighted that AI-assisted cancer contouring provided predictions of tumor size that were 45 times more accurate compared to those made without AI support. Shyam Natarajan, an assistant adjunct professor at UCLA and the senior author of the study, remarked that AI assistance resulted in greater accuracy and consistency among doctors.
Typically, doctors rely on MRIs to evaluate tumor sizes; however, some tumors remain “MRI-invisible,” as noted by Dr. Wayne Brisbane, an assistant professor at the David Geffen School of Medicine at UCLA. According to him, AI proves beneficial in instances where MRIs fall short.
Dr. Brisbane emphasized that leveraging AI in cancer treatment could enhance patient care by providing more personalized and successful treatment strategies. He expressed confidence that AI could exceed human capabilities in certain context.
Avenda Health’s CEO, Dr. Shyam Natarajan, expressed optimism regarding the validation of their innovation through research and acknowledgment by the American Medical Association.
As reported by the American Cancer Society, approximately one in eight men in the U.S. will be diagnosed with prostate cancer in their lifetimes, with one in 44 facing a fatal outcome. This year alone, it is estimated that there will be 299,010 new prostate cancer cases in the U.S., leading to 35,250 deaths from the disease.