November 11, 2022 – Artificial intelligence has a lot of potential in medicine, helping doctors detect skin cancer, report potential problems with chest X-rays, and assist with many other procedures. Screening for colorectal cancer during a colonoscopy is another prime example.
Colonoscopy — recommended for Americans with an average risk of developing cancer at age 45 — isn’t much different for patients with AI or AI. But behind the scenes, AI may make detecting any precancerous growths or lesions more likely.
“AI-enhanced colonoscopy is effectively supercharging a clinician’s ability to find the most accurate of precancerous polyps,” says Tyler M. Berzin, MD, a gastroenterologist at the Center for Advanced Endoscopy at Beth Israel Deaconess Medical Center in Boston.
This technology is designed to mark anything a computer “sees” as suspicious, but it does not replace the training and experience of a gastroenterologist. Even using artificial intelligence, doctors remain by the patient’s side and perform the procedure.
The doctor is still quite in control, says Prateek Sharma, MD, a gastroenterologist and professor of medicine at the University of Kansas School of Medicine in Kansas City. “Artificial intelligence helps them and alerts them to colon polyps – precancerous lesions in the colon – so a doctor can remove them.”
Size, height and numbers are important with polyps. Doctors generally remove or biopsy lesions that are 10 mm or larger in diameter.
But there is still less consensus about the best way to treat smaller polyps.
The clinical significance of detecting and removing small (5 to 9 mm) or insignificant (<5 mm) adenomas is The subject of ongoing debate,” Berzin and co-authors wrote in the May 2020 leading journal Gastroenterology.
One potential drawback to using AI polyp tools, for example, is “the risk of removing a greater number of scanty or hyperplastic polyps, which increases cost and risk, without any benefit to the patient,” Berzin says.
“Trained gastroenterologists are experts in identifying and removing precancerous colon polyps,” Berzin says. “But a gastroenterologist working with an AI polyp detection tool has a huge advantage because AI computer vision tools can analyze every pixel An endoscopic monitoring device and it can do so without distraction or fatigue for even a split second.”
The benefit to patients is “another pair of eyes looking for adenoids and a doctor’s help,” says Sharma, who is also the chair of the Artificial Intelligence Task Force at the American Society of Gastrointestinal Endoscopy.
How it works
Artificial intelligence relies on computer instructions called algorithms that learn the difference between disturbing and benign colonoscopy images and videos. AI improves on the item in a process called machine learning. When the AI system detects a potential area of concern, the technology draws attention to it by framing it inside a box on the screen. Some systems also emit an audible alarm.
“We are seeing more interest in using these algorithms as they will standardize polyp detection in endoscopic specialists, thus reducing the number of missed colon cancers,” says Sravanthi Parasa, MD, a gastroenterologist at the Swedish Health Services in Seattle.
“These products are slowly gaining traction. While scheduling a colonoscopy, patients should ask if their endoscopy specialist has access to enhanced diagnostic tools,” she says.
This technique is not accurate 100% of the time – there can be false positives as the system flags a bubble in the colon, for example, as potentially dangerous. That’s just one reason doctors still have the final say on whether the appendix is suspicious or not.
Artificial intelligence or lack of AI, “Colonoscopy has always been our most effective tool for preventing colon cancer, and for detecting precancerous polyps earlier than any other screening method,” says Berzin, who is also an assistant professor of medicine at Harvard Medical School. .
AI can be expensive
Artificial intelligence and machine learning are already playing a role in “smart” technologies (smartphones, smart watches, smart speakers), self-driving cars, and speech recognition software. But the use of artificial intelligence in medicine is relatively new. And like a lot of new technologies, it’s also very expensive. “AI equipment has to be bought and it is very expensive,” Sharma says.
Parasa agrees that “the cost of algorithms currently can be prohibitive for some centers in the current healthcare landscape.” “The cost will likely come down as more algorithms enter the GI market, as is the case with other software solutions.”
Colorectal cancer is common
Colorectal cancer is the fourth most common type of cancer in Americans, not counting some types of skin cancer. It is also the fourth leading cause of cancer-related deaths in the United States CDC . Reports. More than 150,000 Americans will be diagnosed with colorectal cancer and more than 50,000 will die in 2022, according to National Cancer Institute numbers.
Berzin says more research is needed to examine how humans and this technology interact. “The most interesting research in this field will not be about comparing ‘doctor versus AI’, but it will focus on understanding the nuances of ‘doctor plus AI’. “
In the United States, there are at least three FDA-approved artificial intelligence algorithms for detecting polyps, and more are being developed, Barrassa says.
“In addition, other applications currently available in the European market may be available in the US market in the near future, including characterization of polyps.”
“As the field matures, we are likely to see more and more AI tools that will help us detect and diagnose GI conditions in real time,” she adds. “A combination of algorithms like this will certainly improve patient care and outcomes.”
Although AI is a type of work in progress in medicine, Berzin predicts that the combination of physician technology and AI will “translate into the highest possible protection against colon cancer in the long term.”