Nov. 11, 2022 – Artificial intelligence has great potential in medicine, helping doctors find skin cancer, flag potential problems on X-rays and assist in many other procedures. Screening for colon cancer during a colonoscopy is another good example.
Colonoscopy — recommended for Americans at average risk of cancer starting at age 45 — won’t be much different for patients who add AI or AI. But behind the scenes, AI may be more likely to detect any precancerous polyps or cancerous lesions.
“Intelligence-enhanced colonoscopy effectively supercharges the physician’s ability to find even the subtlest precancerous polyps,” says Tyler M. Berzin, MD, a gastroenterologist at the Center for Advanced Endoscopy at Beth Israel Deaconess Medical Center in Boston.
The technology is designed to flag anything the computer “sees” as suspicious, but it does not replace the training and expertise of a gastroenterologist. Even with artificial intelligence, doctors remain at the patient’s side and perform the procedure.
The doctor is in complete control, says Prateek Sharma, MD, a gastroenterologist and professor of medicine at the University of Kansas School of Medicine in Kansas City, KS. “AI is assisting and alerting them to colonic polyps – the pre-stage lesions in the colon – so that the doctor can remove them.
The dispute continues
Size, height and number matter with polyps. Doctors generally remove or remove tissue lesions that are 10 mm and larger.
But there is even less consensus on the best approach for smaller polyps.
“The clinical importance of detecting and removing small (5 to 9 mm) or smaller (less than 5 mm) adenomas is topic for ongoing discussion,” Berzin and co-authors wrote in a leading journal of gastroenterology in May 2020.
One potential downside to using an AI polyp tool, for example, is “the risk of removing a larger number of smaller or hyperplastic polyps, which increases cost and risk, without any benefit to the patient,” says Berzin.
“Trained gastroenterologists are experts in diagnosing and removing polyps in the colon,” says Berzin. “But a gastroenterologist working with an AI polyp analyzer has a big advantage because AI computer vision tools can simultaneously detect each pixel off the mirror screen and can do so without being distracted or fatigued for even a millisecond.”
The benefit for patients is “another pair of eyes looking for polyps and helping the doctor,” says Sharma, who also chairs the artificial intelligence project at the American Society for Gastrointestinal Endoscopy.
How it works
AI relies on computer instructions called algorithms that learn the difference between worrisome and benign colonoscopy images and videos. Artificial intelligence gets better at things in a process called machine learning. When an AI system spots a potential concern, the technology calls attention to it by framing it in a box on the screen. Some systems also beep.
“We see more interest in using these algorithms as they will standardize endoscopist polyp detection and therefore reduce the number of missed colon cancers,” says Sravanthi Parasa, MD, a gastroenterologist at Swedish Health Services in Seattle.
“These products are slowly taking the weather. When scheduling a colonoscopy, patients should ask if the endoscopist has access to advanced diagnostic equipment,” she says.
The technology isn’t accurate 100% of the time – there can be false positives, with the system flagging an abscess in the colon, for example, as potentially dangerous. That’s just one reason why doctors still have the final say on whether polyps are suspicious or not.
AI or no AI, “colonoscopy has long been our most effective tool for colon cancer prevention, detecting early-stage polyps earlier than any other screening method,” says Berzin, who is also an associate professor of medicine at Harvard Medical School.
AI can be expensive
AI and machine learning already play a role in “smart” technology (smartphones, smartwatches, and smart speakers), self-driving cars, and speech recognition software. But the use of artificial intelligence in medicine is relatively new. And like many new technologies, it is also expensive. “The AI equipment has to be purchased and is expensive,” says Sharma.
“The cost of the algorithms currently may be prohibitive for some facilities in the current healthcare landscape,” agrees Parasa. “The cost is likely to decrease as more algorithms enter the GI market, as with other software solutions.”
Colon cancer is common
Excluding some types of skin cancer, colon cancer is the fourth most common cancer in the United States. It is also the fourth leading cause of cancer-related deaths in the United States, the CDC reports. More than 150,000 Americans will be diagnosed with colon cancer and more than 50,000 will die by 2022, according to the Cancer Institute figures.
More research is needed to explore how humans and this technology interact, says Berzin. “The most interesting research in this area will not be about comparing ‘doctor vs AI’, but will focus on understanding the nuances of ‘doctor vs AI’.”
In the United States, there are at least three FDA-approved artificial intelligence algorithms for detecting polyps, and more are in development, Parasa says.
“In addition, other applications currently available in the European market may be available in the US market in the near future, including polyp symptoms.”
“As the field matures, we will likely see more AI-enhanced tools that will assist us in diagnosing and diagnosing gastrointestinal diseases in real time,” she adds. “A suite of algorithms like this will surely improve patient care and outcomes.”
Even though artificial intelligence is already in place in medicine to some extent, Berzin expects that the combination of medical and AI technology “will result in the highest possible protection against colon cancer in the long term.”