AIDEN: Combination of Artificial Intelligence and Mucosal Exposure Device to Enhance Colorectal Neoplasia Detection
Study Details
Study Description
Brief Summary
The investigators hypothesize that the combined use of CADe system (ENDOAID) and mucosal exposure device (Endocuff Vision®) would improve the adenoma detection rate when compared to CADe system alone.
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
N/A |
Detailed Description
Colorectal cancer (CRC) is the second most common and lethal cancer in Hong Kong with more than 5,500 new cases and 2,200 deaths annually, accounting for 15.8% and 15.4% of all cancer incidence and deaths respectively, according to the Hong Kong Cancer Registry in 2019. (1) The colorectal adenoma-carcinoma sequence has been recognized to be the major pathogeneis of CRC development. (2) Colonoscopy screening has been shown to reduce CRC-related mortality, through removing pre-malignant polyps or detecting early stage CRC. (3)
However, high polyp miss rates were reported in standard colonoscopies, up to 26% for adenomas and 9% for advanced adenomas. (4) As a result, a significant proportion of interval CRC was attributed to the missed lesions during index colonoscopy, leading to adverse outcomes and mortality. (5,6) Risk factors of missed lesions included proximal location, serrate or flat morphology, poor bowel preparation, short withdrawal time and endoscopist experience. (7-9)
To overcome this pitfall, various methods were developed to improve adenoma detection rate (ADR), which was one of the most important quality indicators for colonoscopies. It was proven to be inversely associated with the risks of interval colorectal cancer and fatality. For each 1% increase in the ADR, it was associated with a 3% decrease in the risk of interval CRC and 5% decrease in fatal interval CRC. (10) Techniques including water exchange method (11), second examination of the right colon (retroflexion or second forward view) (12,13) and distal attachment devices (14) were shown to increase ADR. Among the mucosal exposure devices, Endocuff Vision® was extensively evaluated in randomized controlled studies and shown to be superior than conventional colonoscopies with an ADR gain by 4.7-9.8%. (15-18)
Recently, artificial intelligence and automatic computer-aided polyp detection (CADe) systems have developed rapidly and revolutionized the medical field. A number of randomized controlled trials reported a significant benefit in colorectal neoplasia detection with the use of CADe over standard colonoscopy. (19-23) The ADR and number of adenoma detected per colonoscopy was consistently higher among different polyp sizes, locations and morphologies. (24,25) Nevertheless, there was a potential limitation of 'blind spot' if the endoscopist failed to expose the colonic mucosa adequately leading to missed lesions. To date, no available data or ongoing clinical trial was reported regarding the combined use of CADe and mucosal exposure devices. It may enhance the performance of CADe and provide a novel definitive solution to this unaddressed yet important clinical problem.
In this prospective parallel randomized controlled trial, we aim to evaluate the impact of combined use of a new CADe system (ENDOAID) and mucosal exposure device (Endocuff Vision®) on colorectal neoplasia detection.
Study Design
Arms and Interventions
| Arm | Intervention/Treatment |
|---|---|
| No Intervention: ENDOAID CADe system will be used during withdrawal phase of colonoscopy. |
|
| Experimental: ENDOAID with ENDOCUFF CADe system and mucosal exposure device (ENDOCUFF) will be used during withdrawal phase of colonoscopy. |
Device: ENDOCUFF
ENDOCUFF mucousal exposure device will be used during withdrawal process of the colonoscopy.
|
Outcome Measures
Primary Outcome Measures
- ADR [During the colonoscopy]
adenoma detection rate
Secondary Outcome Measures
- ADR for adenomas of different sizes [During the colonoscopy]
<5mm, 5-10mm, >10mm
- ADR for adenomas of different colonic segments [During the colonoscopy]
caecum, ascending colon, hepatic flexure, transverse colon, splenic flexure, descending colon, sigmoid colon, rectum
- Mean number of adenomas per colonoscopy [During the colonoscopy]
Mean number of adenomas per colonoscopy
- Advanced adenoma detection rate [During the colonoscopy]
Advanced adenoma detection rate
- Sessile serrated lesion (SSL) detection rate [During the colonoscopy]
Sessile serrated lesion (SSL) detection rate
- Polyp detection rate [During the colonoscopy]
Polyp detection rate
- Non-neoplastic resection rate [During the colonoscopy]
defined as absence of adenoma or SSL or cancer within resected specimen
- Caecal intubation time [During the colonoscopy]
Caecal intubation time
- Withdrawal time [During the colonoscopy]
excluding interventions
- Total procedural time [During the colonoscopy]
Total procedural time
- False positive rate [During the colonoscopy]
defined as computer artifacts due to colonic mucosal wall or bowel content lasting for >2 seconds
Eligibility Criteria
Criteria
Inclusion Criteria:
-
Aged 18 years old or above;
-
They require elective colonoscopy for colorectal cancer screening, polyp surveillance, or investigation of symptoms such as anemia or altered bowel habit;
-
Written informed consent obtained.
Exclusion Criteria:
-
Contraindication to colonoscopy (e.g. intestinal obstruction or perforation)
-
Contraindication or conditions precluding polyp resection (e.g. active gastrointestinal bleeding, uninterrupted anticoagulation or dual antiplatelets)
-
Staged procedure for polypectomy or biopsy of known unresected lesions
-
Previous surgical resection of colon
-
Personal history of colorectal cancer
-
Personal history of familial polyposis syndrome
-
Personal history of inflammatory bowel disease
-
Advanced comorbid conditions (defined as American Society of Anesthesiologists grade 4 or above)
-
Pregnancy
-
Unable to obtain informed consent
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- Chinese University of Hong Kong
Investigators
None specified.Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- CREC2022.152