| Background and Aim:Colorectal cancer(CRC)is one of the common cancers.CRC originates from adenomatous polyps(or adenomas).Although adenomas are initially benign,they may become cancerous and spread to adjacent and distant organs over time.Improving the rate of polyp detection is an important measure to prevent colorectal cancer.A real-time automatic polyp detection system,which performance is close to that of expert endoscopists,can help endoscopists to examine polyps.Real-time automatic polyp detection systems,through deep learning methods,can learn and perform specific endoscopic tasks previously performed by endoscopists.The purpose of this study was to explore whether a high-performance,real-time automatic polyp detection system can improve the polyp detection rate(PDR)in the actual clinical environment.Methods:This study will include patients who will undergo enteroscopy in the Digestive Endoscope Center of Nanfang Hospital from August 2019 to December 2019.The selected patients underwent same-day,back-to-back colonoscopies in a random order,with either traditional colonoscopy or artificial intelligence(AI)-assisted colonoscopy performed first by different experienced endoscopists(>3000 colonoscopies).The system displayed the position of the detected polyp with a hollow blue tracking box on the adjacent monitor.The primary outcome was the PDR.Results:In the study,we randomized 150 patients.The AI system significantly increased the PDR(34.0%vs 38.7%,p<0.001).In addition,Artificial intelligence-assisted colonoscopy improved the detection rate of polyps less than 6 mm(69 vs 91,p<0.001),but no difference was found with regard to larger lesions.Artificial intelligence-assisted colonoscopy also improved the detection rate of Paris type 0-Ⅱa polyps(26%vs 32.0%,p<0.001).There was a total of 52 false positives in the AI-assisted colonoscopy group,averaging 0.35 false positives per colonoscopy,mostly due to feces,mucosal folds and so on.Conclusions:Compared with traditional colonoscopy,AI-assisted colonoscopy can significantly increase the number of polyps and the PDR.AI-assisted colonoscopy can significantly improve the detection of diminutive polyps(diameter<6mm)as well as Paris type 0-IIa polyps.In view of its high accuracy,fidelity and stability,the current CADe system may be used in current clinical practice to better detect colorectal polyps.Artificial intelligence will be an important part of doctors’diagnosis of patients,selection of treatment methods and prediction of results. |
