An artificial intelligence healthcare company claims its software can detect the extent of prostate cancer with greater accuracy than human doctors.
Avenda Health published a study last month involving ten doctors who assessed 50 different cases of prostate cancer. The company’s Unfold AI software achieved an accuracy rate of 84.7% in detecting cancer, compared to the 67.2% to 75.9% accuracy range of the physicians who conducted manual detections.
Conducted in partnership with UCLA Health and featured in the Journal of Urology, the study also indicated that the application of AI in cancer contouring resulted in predictions of cancer size being 45 times more precise and consistent than without AI assistance.
Shyam Natarajan, an assistant adjunct professor of urology, surgery, and bioengineering at UCLA and senior author of the study, stated that the integration of AI support made physicians more accurate and consistent, leading to a higher degree of agreement among them.
Typically, doctors rely on MRI scans to determine tumor sizes, but some tumors can be “MRI-invisible,” noted Dr. Wayne Brisbane, an assistant professor of urology at the David Geffen School of Medicine at UCLA. He emphasized the crucial role of AI in situations where MRIs are inadequate.
Brisbane mentioned that the implementation of AI in cancer treatment could result in more personalized and effective care for patients, tailoring treatments to meet individual needs more successfully in combatting the disease. He added that AI has the potential to “go beyond human ability.”
Avenda Health CEO Dr. Shyam Natarajan expressed that it is “empowering for physicians” to see such innovations validated through research and recognized by the American Medical Association.
According to the American Cancer Society, approximately 1 in 8 men will receive a prostate cancer diagnosis in their lifetime, and 1 in 44 men will succumb to the disease. The organization estimates that there will be 299,010 new cases of prostate cancer in the US this year, with 35,250 deaths attributed to the illness.