The researchers have managed to develop software that seems to be capable of detecting bowel cancer in less than a second or so they claim. During their trials, the AI appeared to be successful at being able to spot colorectal adenomas, which are basically benign tumors that have the potential to evolve into cancer, from magnified endoscopic images.
The system was trained by matching these images against 30,000 others during the machine learning process, thus allowing it to quickly identify what could be cancer and what might not. Presenting the results at the United European Gastroenterology Week in Barcelona, Spain, study leader Dr Yuichi Mori from Showa University in Yokohama believes that this could save patients from having needless surgery.
Mori was quoted as saying, “This allows the complete resection of adenomatous (cancerous) polyps and prevents unnecessary polypectomy (removal) of non-neoplastic polyps.” However as it stands the system has yet to be approved by the country’s regulators, but Mori believes that the results make it acceptable for clinic application.