In recent years, with the continuous development of artificial intelligence, automatic polyp detection systems have shown its potential in increasing the colorectal lesions. Yet, whether this system can increase polyp and adenoma detection rates in the real clinical setting is still need to be proved. The primary objective of this study is to examine whether a combination of colonoscopy and a deep learning-based automatic polyp detection system is a feasible way to increase adenoma detection rate compared to standard colonoscopy.

Inclusion Criteria:

• Patients aged between 40-85 years old who have indications for screening, surveillance and diagnostic.
• Patients who have signed inform consent form.

Exclusion Criteria:

• Patients who have undergone colonic resection
• Patients with intracranial and/or central nervous system disease, including cerebral infarction and cerebral hemorrhage.
• Patients with severe chronic cardiopulmonary and renal disease.
• Patients who are unwilling or unable to consent.
• Patients who are not suitable for colonoscopy
• Patients who received urgent or therapeutic colonoscopy
• Patients with pregnancy, inflammatory bowel disease, polyposis of colon, colorectal cancer, or intestinal obstruction
• Patients who are taking aspirin, clopidogrel or other anticoagulants
• Patients with withdrawal time < 6 min