An artificial intelligence healthcare company claims its software is capable of identifying the progression of prostate cancer with greater accuracy than doctors.
Avenda Health conducted a study last month involving ten physicians, each evaluating 50 cases of prostate cancer. Their Unfold AI software accomplished cancer detection with an accuracy rate of 84.7%, whereas the physicians’ manual assessments ranged from 67.2% to 75.9%.
This study, which was carried out in collaboration with UCLA Health and published in the Journal of Urology, revealed that the use of AI for cancer contouring resulted in predictions of cancer size being 45 times more accurate and consistent compared to assessments without AI support.
According to Shyam Natarajan, a senior author of the study and an assistant adjunct professor at UCLA, the implementation of AI assistance enhanced both the accuracy and consistency of the doctors, leading to a greater agreement among them during evaluations.
Doctors typically rely on MRIs to assess tumor sizes; however, some tumors are “MRI-invisible,” explained Dr. Wayne Brisbane, an assistant professor of urology at UCLA’s David Geffen School of Medicine. He noted that AI addresses the shortcomings of MRIs.
Dr. Brisbane emphasized that the application of AI in cancer treatment could pave the way for more effective and personalized patient care, with therapies tailored to individual needs and greater success in combating the disease. He stated that AI has the potential to surpass human capabilities.
Avenda Health’s CEO, Dr. Shyam Natarajan, expressed that it is encouraging for physicians to witness such innovations being validated by studies and acknowledged by the American Medical Association.
Statistics from the American Cancer Society indicate that approximately 1 in 8 men in the U.S. will receive a prostate cancer diagnosis in their lifetime, and 1 in 44 men will succumb to the illness. This year, it is projected that there will be 299,010 new prostate cancer cases in the U.S., with 35,250 fatalities attributed to the disease.