An appendix is a small growth in colon tissue that can develop into colorectal cancer. Patients with polyps have a higher risk of developing cancer in the future.
The use of artificial intelligence (AI) during colonoscopy, and endoscopic camera examination, could be useful for detecting precancerous polyps, says Yoichi Mori, MD, associate professor of the Clinical Efficacy Research Group at the University of Oslo.
The artificial intelligence tool used during colonoscopy is based on the same facial recognition technology as on mobile phones. By marking polyps in red or green squares, the tool assists doctors during a colonoscopy, helping them see polyps they would otherwise have missed.
Morey says this technique is useful when polyps are small or difficult to detect.
More patients are advised to have a control colonoscopy every three years
A potential drawback of the new technology is that doctors discover more small, non-hazardous polyps with the instrument, which must then be removed. Even if removing small polyps doesn’t have any bigger side effects, about one in every thousand will experience a smaller colon bleed or rupture during removal.
Morey and colleagues compared results after screening for colorectal cancer, with and without AI. About 5,800 patients were included in the study.
Morey says the use of AI tools during colonoscopy has increased the number of polyp patients recommended to be screened every three years by 20 percent in Europe. Patients are usually advised to have controls every 10 years.
The research is published in the effect of artificial intelligence on colonoscopy monitoring after polypectomy: a pooled analysis of randomized trials.
Further examination can lead to overdiagnosis
As patients have to undergo more testing, the use of AI during colonoscopy can lead to overdiagnosis, stress on patients, and higher financial costs for health services. Some patients also find colonoscopy uncomfortable. Taking laxatives and drinking two to three liters of clear fluids before the exam can be difficult.
We don’t know whether more colonoscopies will result in a lower risk of colorectal cancer over time, or whether it will only increase financial pressure on hospitals. This requires longitudinal clinical efficacy studies and studies of the financial costs of using AI during colonoscopy.”
Yuichi Mori, Physician and Associate Professor, Clinical Efficacy Research Group, University of Oslo
Lower costs when using AI during screening
In another new article, Morey and colleagues examined whether the use of AI during colorectal cancer screening actually leads to higher financial costs for health services. The study showed that using AI tools during colorectal cancer screening in the United States reduced costs by $57 per person.
Using AI during screening could prevent approximately 7,200 cases of colorectal cancer and approximately 2,000 related deaths in the US population. This will save society costs of $290 million annually in the long run. The study is based on modeling, and we don’t yet know if the estimates are correct. To be able to tell, we’ll need large-scale clinical studies, Morey says.
The research is published in Artificial Intelligence Cost-Effectiveness of Colonoscopy Screening: A Modeling Study.
Morey is now working on the EU-funded Oprah Research Project, which is a set of large-scale clinical studies of the long-term effects of using artificial intelligence during screening.
The goal is to change the standard of treatment by establishing reliable clinical evidence, and in the long term to improve outcomes for patients, for example when colorectal cancer is detected. Morey says Oprah’s project will find the answers to these important questions.
New program for screening for colorectal cancer in Norway
Norway recently introduced a national screening program for colorectal cancer. Screening for 55-year-old men and women through colonoscopy or stool samples will be offered starting this fall.
– I am pleased that Norway now offers a colorectal cancer screening program, and I hope it will contribute to reducing cancer, says Morey.
– We don’t know yet if AI can contribute to making a difference during screening, but I hope it will be used when it’s useful, he finished.