Michael Wallace has it He has performed hundreds of colonoscopies in his 20 years as a gastroenterologist. He believes he is very good at identifying benign tumors that can appear along the ridges of the colon and possibly turn into cancer. But it isn’t always perfect. Sometimes the polyps are flat and difficult to see. Other times, doctors miss them. “We are all human,” says Wallace, who works at Mayo Clinic. After a morning of cascading actions requiring attention to minute details, he says, “We are tired.”
Colonoscopy, if bothersome, it is Very effective In the detection of precancerous polyps and the prevention of colon cancer. But the effectiveness of the procedure largely depends on the capabilities of the doctor who performs it. Now, the Food and Drug Administration has approved a new tool that promises to help doctors recognize precancerous growths during colonoscopy: Artificial intelligence System made by Medtronic. Doctors say that along with other measures, the tool can help improve the prognosis. “We really have a chance to completely eliminate colon cancer in anyone who gets screened,” says Wallace, who consulted with Medtronic on the project.
The Medtronic system, called GI Genius, has seen more colon from the inside than most clinicians. Medtronic and its partner Cosmo Pharmaceuticals trained the algorithm to recognize polyps by reviewing more than 13 million videos of colonoscopies performed in Europe and the US that Cosmo collected while conducting drug trials. In order to “teach” the AI to distinguish between potentially dangerous growths, the images were classified by gastroenterologists as either normal or unhealthy tissues. Then the AI was tested on polyps that were difficult to recognize gradually, ranging from colonoscopy performed under ideal conditions and moving on to more difficult challenges, such as distinguishing a tumor that was very small, only within the scope of the camera for a while, or hidden in a dark spot. .
The system, which can be added to the ranges doctors are already using to perform a colonoscopy, follows it while the doctor examines the colon, highlighting potential polyps with a green box. GI Genius was approved in Europe in October 2019 and is the first artificial intelligence approved by the Food and Drug Administration to help detect polyps in the colon and rectum. Says Wallace, who co-authored First verification study By GI Genius. “It’s an impressive system.”
Mark Pocaben, a gastroenterologist at NYU Langone, who was not involved in creating GI Genius, says it makes sense that AI would be good at identifying polyps. “There is less diversity when you look at polyps,” says Pochapin. Millions of colonoscopy videos provide a lot of data to make the algorithm comprehensive. It should protect the system from worries Bias in other healthcare algorithms. “There are only too many types of polyps,” he says.
Medtronic sees GI Genius and other AI tools as the cornerstone of its future business, says Giovanni Di Napoli, Medtronic GI’s business head. To this end, the company has invested a lot of time and resources to win Food and Drug Administration approval for this device. “It took nearly a year for us to get approval from the Food and Drug Administration,” says Di Napoli. “It is not easy.”
Medtronic sought approval from the FDA under what the agency calls its de novo pathway, which requires applicants to provide information about the safety and efficacy of the new devices including clinical data. This is a longer, more complex application that some other AI medical devices have avoided. Most medical machine learning and artificial intelligence devices are brought to the market with a simplified FDA app known as the 510 (k) path, which only requires them to prove that their devices are similar to other tools already in use and usually takes about six months. to me A study published in The scalpelAnd the Of the 222 AI devices that were introduced to the market in the US between 2015 and 2020, 92 percent did so with 510 (k).