The new endoscopic system with artificial intelligence automatically detects colorectal adenomas during colonoscopy. The system developed in Japan, has recently been tested in one of the first studies of endoscopic surgery with artificial intelligence in a clinical setting. The results presented at the 25th United European Gastroenterology Week in Barcelona, Spain.
Endotsitoskopiya with artificial intelligence: how it works
The new system uses an automated diagnostic endotsitoskopicheskoe 500-fold enlarged image 300 for analyzing characteristics of the polyp after applying method narrowband imaging or staining with methylene blue. It compares the characteristics of each polyp with over 30,000 endotsitoskopicheskimi images that are used for machine learning, allowing to predict the pathology of less than a second. Preliminary studies have demonstrated the possibility of using such a system for classification of colorectal polyps.
A new study in general practice
In a study conducted by Dr. Yuichi Mori Showa University in Yokohama, Japan, was attended by 250 men and women who have found colorectal polyps using endotsitoskopii. A system supporting an artificial intelligence, has been used to predict each polyp pathology and these readings were compared with pathological report obtained from resected polyp obraztsov.306 assessed in real time using a system with an artificial intelligence, providing 94% sensitivity, 79% specificity and 86 % accuracy.
Speaking at a meeting of the United European Gastroenterology Week, Dr. Mori said: "The most remarkable breakthrough lies in the fact that artificial intelligence allows the optical biopsy of colorectal polyps in real time during a colonoscopy, which allows you to perform a resection of adenomatous polyps and prevent unnecessary poliektomiyu. these results are acceptable for clinical use, and our immediate goal - to get the approval of the regulatory authorities. "