Colon Cancer Detection Rate Drops with AI Use, Study Finds
New research published in the Lancet Gastroenterology and Hepatology journal suggests that the use of artificial intelligence (AI) in colonoscopies may lead to a significant decrease in tumor detection rates. The study, conducted in four Polish endoscopy centers, found that the adenoma detection rate (ADR), which measures the ability to find tumors, dropped from 28.4% to 22.4% after AI tools were introduced.
Researchers analyzed the impact of AI tools on endoscopists’ performance using data from the ACCEPT (Artificial Intelligence in Colonoscopy for Cancer Prevention) trial, which involved four Polish endoscopy centers. The trial introduced AI tools for polyp detection at the end of 2021. The study tracked colonoscopies administered with and without AI assistance, comparing the identification of tumors (ADR) before and after AI implementation.
From September 2021 to March 2022, 1,443 patients underwent non-AI-assisted colonoscopies before and after the introduction of AI. The researchers found that the tumor detection rate decreased significantly, going from 28.4% to 22.4%. The findings highlight a potential drawback of over-reliance on AI in medical procedures, suggesting that continuous exposure to AI might reduce the effectiveness of standard non-AI-assisted colonoscopies.
The researchers concluded that AI exposure, along with patient demographics, was an independent factor associated with the tumor identification rate. Dr. Harvey Castro, a specialist in AI and healthcare, called the study an important piece of work, noting that while AI can aid in cancer detection, it may weaken doctors’ ability to recognize tumors when the tool is not available.
Castro emphasized that even a 1% change in detection rates can significantly impact colon cancer survival for thousands of patients. He recommended integrating AI wisely rather than avoiding it altogether, stressing the importance of balancing technology with human expertise in healthcare.
As AI continues to shape modern medicine, the study underscores the need for careful implementation and oversight. Experts suggest that a balanced approach, combining the strengths of AI with the irreplaceable skills of human clinicians, is essential for ensuring safe and effective patient care. The findings serve as a reminder that while technology can enhance medical practices, it should not replace the critical judgment and expertise of healthcare professionals.
“,”category”:”health”,”tags”:[“colon cancer”,”AI in healthcare”,”tumor detection”,”medical technology”,”diagnostic accuracy”,”adenoma detection rate”,”AI impact on medical diagnostics”],”has_financial_effect”:false,”political_stance”:”neutral”}