An intelligent computer program that can pinpoint bowel cancer in less than a second could be used to diagnose patients, experts believe.
In a trial, the AI system was able to identify potentially dangerous tumours from endoscopy images with clinical accuracy.
Scientists used the program to assess 306 colorectal polyps - growths in the bowel which may or may not be cancerous - in 250 men and women.
It took less than a second for the system to analyse each magnified endoscopic image and decide whether or not the polyp was malignant.
The program correctly picked up 94 per cent of cancers by matching each growth against more than 30,000 images that were used for machine learning.
Study leader Dr Yuichi Mori from Showa University in Yokohama, Japan, said: "The most remarkable breakthrough with this system is that artificial intelligence enables real-time optical biopsy of colorectal polyps during colonoscopy, regardless of the endoscopists' skill.
"This allows the complete resection of adenomatous (cancerous) polyps and prevents unnecessary polypectomy (removal) of non-neoplastic polyps.
"We believe these results are acceptable for clinical application and our immediate goal is to obtain regulatory approval for the diagnostic system."
The results were presented at United European Gastroenterology (UEG) Week in Barcelona, Spain.
