Adenoma Detection Rate in Water and Air Colonoscopy Using Computer-aided System

Adenoma Detection Using Real-Time Computer-Aided Colon Polyp Detection System to Compare Water Exchange and Air Insufflation - A Randomized Controlled Trial

Water exchange (WE) is an effective insertion method to minimize insertion discomfort and maximize ADR. It is characterized by infusing water to guide insertion in an airless lumen with suctioning of infused water during insertion and almost complete removal of the infused water when cecal intubation is achieved. A modified Delphi review reported water exchange showed the highest overall ADR, ADR in screening cases, and in the right side of the colon compared with water immersion and air (or CO2) insufflation. One of the plausible mechanisms of improving ADR by water exchange is salvage cleaning during insertion, which might help artificial intelligence by removing the interference of fecal debris and bubbles. However, no RCT has been performed to evaluate the effect of CADe on WE colonoscopy. Therefore, investigators will conduct a RCT comparing the ADR of CADe assisted colonoscopy inserted with either WE or insufflation. Our hypothesis will be that CADe assisted WE colonoscopy achieves higher ADR than CADe assisted air insufflation colonoscopy.

The primary end point will be ADR, defined as the proportion of patients with at least one adenoma. Serrated lesions will not be counted as part of ADR. Secondary outcomes were FP rate, adenomas per colonoscopy (APC), polyp detection rate, polyps per colonoscopy, SL detection rate, cecal intubation rate, withdrawal time. Missed polyp will be defined as a polyp recognized by endoscopist, but is not detected by CADe system at an interval of 2 seconds. The detected polyps will be classified in sizes, morphology (polypoid or nonpolypoid). Investigators also record the detection of every polyp by the endoscopist or the CADe system in time priority. If a polyp is detected by performing endoscopist at the same time as the CADe system, the credit will go to the endoscopist. The colonoscopy procedure will be recorded as CADe over-laid videos and be reviewed for FP characteristics, numbers and appearance time afterward. The withdrawal time will be divided into inspection time, cleaning time and polypectomy time. Boston Bowel Preparation Scale score and the amount of water infused and aspirated will also be recorded.

This group of participants receives water exchange colonoscopy with the computer-aided polyp detection system.

This group of participants receives air insufflation colonoscopy the computer-aided polyp detection system.

Using computer-aided polyp detection algorithm on colonoscopy with water exchange or air insufflation methods. This computer-aided algorithm does not intervene directly the colonoscopy procedure.

proportion of patients with at least one colorectal adenoma detected among all patients examined by an endoscopist

Inclusion Criteria: Consecutive patients, undergoing screen, diagnostic or surveillance colonoscopy performed by the participating endoscopists will be enrolled. Exclusion Criteria: We excluded patients age < 50 or > 80 years old, undergone colonoscopy in the past 3 years, partial intake of bowel preparation, renal failure, previous colonic resection, scheduled for polypectomy, American Society of Anesthesiology (ASA) Risk Class 3 or higher, and lack of written informed consent.