An AI healthcare company has reported that its software demonstrates a greater accuracy in detecting the extent of prostate cancer compared to human doctors.
Avenda Health conducted a study involving ten doctors who evaluated a total of 50 prostate cancer cases. The company’s Unfold AI software achieved an accuracy rate of 84.7% in identifying cancer, while the physicians’ assessments ranged between 67.2% and 75.9%.
The research, a collaboration with UCLA Health and published in the Journal of Urology, indicated that utilizing AI for cancer contouring resulted in predictions regarding tumor size being 45 times more accurate and consistent compared to methods without AI.
According to Shyam Natarajan, an assistant adjunct professor of urology, surgery, and bioengineering at UCLA, the integration of AI assistance improved both the accuracy and consistency of doctors’ assessments, leading to a higher level of agreement among them when AI was involved.
While doctors typically rely on MRIs to evaluate tumor sizes, Dr. Wayne Brisbane, an assistant professor of urology at UCLA’s David Geffen School of Medicine, noted that certain tumors are “MRI-invisible.” He emphasized that AI can provide solutions where MRIs may fall short.
Brisbane remarked that the incorporation of AI in cancer treatment could potentially enhance personalized patient care, allowing for treatment plans that are more specifically tailored and more effective in combating the disease. He highlighted that AI has the capability to surpass human diagnostic abilities.
Avenda Health’s CEO, Dr. Shyam Natarajan, expressed the importance of validating such innovations through studies and receiving recognition from the American Medical Association.
Currently, prostate cancer affects about 1 in 8 men in the United States, with an estimated 1 in 44 men succumbing to the disease, as reported by the American Cancer Society. This year, approximately 299,010 new cases of prostate cancer are projected in the US, with a fatality estimate of 35,250.