Michael Wallace said performed 20-year-old colonoscopies as a gastroenterologist. They are thought to be able to detect growths, or microbes, which can spread to the edges of the cage and possibly to cancer. But he is not always perfect. Sometimes the tiny insects are flat and difficult to spot. At times, doctors simply do not need them. Wallace, who works at the Mayo Clinic, says: “We are all human beings. After a morning of back-and-forth procedures that require attention to detail, he says, “we’re tired.”
Colonoscopies, if not fun, are very helpful in the fight against cancer-causing organisms and in the prevention of colon cancer. But the effectiveness of this procedure depends to a large extent on the skill of the physician. Now, the Food and Drug Administration has approved a new tool that promises to help doctors recognize early growth during colonoscopy: a artificial intelligence a system developed by Medtronic. Doctors say that in addition to other things, the tool can help correct disease. “We have the opportunity to completely eradicate colon cancer in anyone who can be diagnosed,” said Wallace, who consulted Medtronic on the project.
The Medtronic method, called GI Genius, has seen more colonies than most physicians. Medtronic and its affiliates Cosmo Pharmaceuticals trained algorithms to detect polyps by monitoring more than 13 million colonoscopies in Europe and the US where Cosmo collected drug testing. To “train” AI to differentiate potentially dangerous growth, the images were cited by gastroenterologists as healthy or unhealthy muscles. AI was then gradually tested to detect microbes, starting with colonoscopies that were performed in a safe environment and moving on to other problems, such as distinguishing a polyp that was too small, just a small camera, or hiding places. akuda.
The procedure, which can be added to a supplement that doctors already use to perform a colonoscopy, follows a doctor’s examination of the intestines, highlighting potential polyps with a green box. GI Genius was approved in Europe in October 2019 and is the first AI developed by the FDA to help them identify polyps polyps. “It got things that even I miss,” says Wallace, who co-wrote it first study to prove of GI Genius. It’s a fun way. ”
Mark Pochapin, a gastroenterologist at NYU Langone who did not participate in the development of the GI Genius, says it is logical that AI could be a good tool for diagnosing polyps. “There are fewer species compared to other microbes,” Pochapin said. Millions of colonoscopy videos provide more information to make the algorithms more complete. This should protect the system from worries about bias in other health systems. “There are so many types of microbes,” he said.
Medtronic sees GI Genius, and other AI tools, as the cornerstone of the future market, says Giovanni Di Napoli, president of Gt’s Medtronic business. To achieve this, the company wasted time and resources to obtain approval from the FDA for this product. “It took us about a year to get FDA approved,” says Di Napoli. It’s not easy. ”
Medtronic sought FDA approval under the agency’s call for a de novo approach, which required applicants to provide safety information for new devices including more medical information. This is a long and arduous process that some AI medical devices have avoided. Many AI devices and machine learning tools go to market using the simple FDA method called 510 (k) method, which only requires them to ensure that their devices are compatible with other devices already in use and take about six months. According to research published in Lancet, Of the 222 AI devices sold in the US between 2015 and 2020, 92% said through 510 (k).
Source link
