Artificial Intelligence-assisted Colonoscopy With or Without Endocuff Vision

E-DA Hospital

Status: Enrolling

Conditions: Adenoma Detection Rate

Treatments: Device: High-definition endoscope; Device: Endocuff vision; Device: Artificial intelligence

Study type: Interventional

Funder types: Other

Identifiers

NCT05863208

EMRP53109N

Details and patient eligibility

About

Adenoma detection rate (ADR) is considered the single most important quality measure in colonoscopy and a higher ADR can reduce the risk of interval colorectal cancer (CRC). Several kinds of new endoscopes and accessories have been accessed to investigate the abilities of improving the ADR. Artificial intelligence (AI) and Endocuff vision are promising new devices to improve the ADR. However, the effect of combining AI and Endocuff vision on ADR remains unclear. The aim of this prospective randomized study is to compare the ADR of AI plus Endocuff vision, AI alone and standard colonoscopy examination.

Full description

This is a prospective single-blinded randomized controlled trial of three different types of colonoscopy examinations by 1:1:1 ratio. We use EndoAim AI (ASUS, Taiwan) and Endocuff vision (Olympus, UK) assisted colonoscopy in the first group. We use AI assisted colonoscopy in the 2nd group. We use standard colonoscopy in the 3rd group.

Eligible patients are older than 40 years old and receive colonoscopy for either symptomatic or screening/surveillance. All endoscopists should receive training on EndoAim AI systems and Endocuff vision. During the procedure, experienced endoscopists use high-definition endoscopes (EVIS-EXERA 290 video system, Olympus Optical, Aizu, Japan) under white light and insert to the cecum in the three different groups. The cecal intubation is confirmed by the identification of ileocecal valve and appendiceal orifice.

The Boston Bowel Preparation Scale is used for grading the bowel preparation quality. The size (compared with biopsy forceps), location and morphology of polyps are recorded by the independent endoscopist. All polyps ae removed by either biopsy or polypectomy. The insertion and withdrawal time are measured. The time of the polypectomy site is not included in the withdrawal time.

Enrollment: 1,000 estimated patients

Sex: All

Ages: 40+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

Patients over 20 years old are undergoing outpatient sedative colonoscopy in the E-Da Hospital, E-Da cancer Hospital and Chung Shan Medical University Hospital in Taiwan

Exclusion criteria

• A prior history of of inflammatory bowel disease, colorectal cancer, previous bowel resection, Peutz-Jeghers syndrome, familial adenomatous polyposis or other polyposis syndromes
• Bleeding tendency
• For scheduled endoscopic treatment

Trial design

Primary purpose: Diagnostic

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: Single Blind

---

1,000 participants in 3 patient groups

Artificial Intelligence-assisted Colonoscopy with Endocuff Vision

Active Comparator group

Description:

Use artificial intelligence-assisted colonoscopy with Endocuff Vision

Treatment:

Device: Artificial intelligence

Device: Endocuff vision

Device: High-definition endoscope

Artificial Intelligence-assisted Colonoscopy

Active Comparator group

Description:

Use artificial intelligence-assisted colonoscopy alone

Treatment:

Device: Artificial intelligence

Device: High-definition endoscope

Standard colonoscopy

Sham Comparator group

Description:

Use standard colonoscopy without artificial intelligence-assisted colonoscopy or Endocuff Vision

Treatment:

Device: High-definition endoscope