Computer-Aided Water Exchange Colonoscopy With and Without Linked-Color Imaging for Detection of Clinically Significant Serrated Lesions

Colon Adenomas, Serrated Lesion, Sessile Colonic Polyp, Sessile Serrated Lesion

The goal of this clinical trial is to compare the detection rate of clinically significant serrated lesions (CSSL) in participants undergoing water exchange (WE) colonoscopy with artificial intelligence (AI)-based computer-aided detection (CADe) for screening, surveillance, diagnosis for symptoms, or referred owing to a positive fecal immunochemical test (FIT) or guaiac fecal occult blood test (gFOBT) result. There will be two arms in this study: WE with AI-assisted CADe (WEAID) control and WEAID plus linked-color imaging (LCI). The main question it aims to answer is whether the addition of LCI into WEAID colonoscopy increases CSSL detection rate. Both groups use water instead of air to insert the colonoscope into the cecum. The control method uses CADe to help detect colonic lesions. The study method uses a combination of CADe and LCI to detect lesions. Researchers will compare CSSL detection rate to see if the addition of LCI increases the detection of CSSL during CADe-assisted WE colonoscopy.

This will be a multicenter, unblinded investigators, prospective randomized controlled trial (RCT). Randomization (1:1 to WEAID alone and WEAID plus LCI) will be based on computer generated random numbers placed inside opaque sealed envelopes. The envelope (in pre-arranged order) will be opened to reveal the code when the colonoscopist is ready to insert the endoscope to begin the examination. This will be a comparison of two different methods with two arms (WEAID, WEAID-LCI) to see which one is better at detecting CSSL. Patient recruitment will be conducted at 4 hospitals: University of Montreal Medical Center (CHUM), Quebec, Canada; Evergreen General Hospital, Taoyuan, Taiwan; King Chulalongkorn Memorial Hospital, Bangkok, Thailand; and Ospedale Valduce, Como, Italy. The study period is expected to be 2 years (from August 2025 to July 2027). Patients aged 40-80 y/o at average risk of colorectal cancer who are willing to participate will sign an informed consent before starting the colonoscopy procedure. Separate parallel randomization will be set up at each site, stratified by investigator and indication of colonoscopy (screening, surveillance, diagnostic, or positive FIT or gFOBT result). Mode of sedation will include no sedation, on-demand sedation, conscious sedation or full sedation with propofol. Randomization will be carried out by computer-generated sequences. The study employs standard high-definition colonoscopy video processors with integrated CADe system (CAD-EYE, EW10-EC02, Fujifilm). The ELUXEO 7000 system (Fujifilm) will be used. The same endoscopist performs the insertion and withdrawal of colonoscopy. The colonoscope will be inserted with WE method, instead of air insufflation. WE involves simultaneous infusion of distilled water to facilitate luminal expansion and suction of unclean water. Control Method: One arm of the study will include CADe as the control method. The CADe system will be activated during the withdrawal phase of the procedures, and it will provide real-time output in the form of a bounding box whenever the CADe device identifies a suspected polyp. Study method: The other arm entails the addition of a commercially available image-enhanced LCI technique (Fujifilm) to the CADe system. LCI will be activated during the withdrawal phase to enhance visualization of the lesions.

No related papers found yet.

Canada, Italy, Taiwan, Thailand