AI-Enhanced Wide-Field Endoscopic Fluorescence Mapping of Gastrointestinal Mucosal Permeability in IBD - A Pilot Study in IBD Patients and Controls

Overview

This pilot study will test a new imaging system that uses fluorescent dye and artificial intelligence (AI) during colonoscopy to measure how "leaky" the lining of the colon is in people with inflammatory bowel disease (IBD). The study will include 70 adults at 3 Canadian hospitals: 60 people with ulcerative colitis or Crohn's disease affecting the colon, and 10 people without IBD who are having colonoscopy for routine colorectal cancer screening or surveillance. During the colonoscopy, participants will receive intravenous fluorescein, and the imaging system will record fluorescence in the colon as the scope is withdrawn. The main goal is to find out whether this method can be used safely during routine colonoscopy and whether it can produce usable measurements of mucosal permeability. The study will also examine whether these measurements are related to standard measures of inflammation seen during endoscopy, in biopsy samples, and in ex vivo Ussing chamber testing at the McMaster site. The control group will help define what normal fluorescence and permeability look like. This study is intended to provide early data on whether this approach could become a useful new way to assess barrier dysfunction in IBD.

Detailed description

This multi-center pilot study will evaluate the feasibility, safety, and clinical relevance of an investigational AI-assisted wide-field fluorescence endoscopy system for assessment of colonic mucosal permeability in inflammatory bowel disease (IBD). The study will enroll 70 adults across 3 Canadian academic centers, including 60 patients with ulcerative colitis or Crohn's disease involving the colon who are undergoing clinically indicated colonoscopy, and 10 non-IBD controls undergoing routine colorectal cancer screening or surveillance.

Impaired intestinal barrier function is an important feature of IBD, but current permeability tests do not provide direct, segment-specific, real-time information during colonoscopy. In this study, standard colonoscopy will be supplemented by intravenous fluorescein and fluorescence imaging during scope withdrawal, together with scope-position tracking and AI-assisted video analysis. The investigational system is designed to generate regional permeability measurements based on fluorescence leakage patterns along the colonic mucosa.

The primary objective is to determine whether this imaging approach can be safely and feasibly integrated into routine colonoscopy and produce interpretable permeability data. Secondary objectives are to evaluate whether fluorescence-derived permeability measurements correlate with regional endoscopic disease activity, matched histopathology, and ex vivo Ussing chamber permeability measurements at the McMaster site. Non-IBD controls are included to provide reference fluorescence and permeability data for non-inflamed mucosa.

This is a non-randomized, observational study in which all participants will undergo the same investigational imaging procedure in addition to standard care. The study is intended to generate preliminary data to support future trials of fluorescence-based permeability mapping as a tool for assessing mucosal barrier dysfunction in IBD.

Interventions

• Diagnostic test Artificial Intelligence-Enhanced Wide-Field Mucosal Permeability Mapping
  This is an observational study, and participants are not assigned a therapeutic intervention as part of the research. All participants will undergo a clinically indicated colonoscopy as part of routine care. As part of the study procedures, colonoscopy will be supplemented with intravenous sodium fluorescein, wide-field fluorescence imaging during scope withdrawal, and AI-assisted video recording and analysis to assess mucosal permeability. In addition, targeted research biopsies will be collect

Primary outcome measures

• • Feasibility of Procedure [Time frame: From enrollment to end of study = one patient visit]
• • Safety of the Imaging System [Time frame: From enrollment to end of study = one patient visit]

Secondary outcome measures (5)

• • Correlation with Endoscopic Severity: [Time frame: From enrollment to end of study = one patient visit]
• • Correlation with Histopathology [Time frame: From enrollment to end of study = one patient visit]
• • Correlation with Ussing ex vivo permeability (Performed at McMaster only) [Time frame: From enrollment to end of study = one patient visit]
• • Accuracy of AI Detection [Time frame: From enrollment to end of study = one patient visit]
• • Inter-Observer Consistency [Time frame: From enrollment to end of study = one patient visit]

Eligibility criteria

Inclusion criteria

• Adults aged 18 years or older
• Able to provide written informed consent
• Established ulcerative colitis or Crohn's disease involving the colon, or non-IBD control undergoing screening/surveillance colonoscopy
• Undergoing clinically indicated colonoscopy
• Ability to comply with study procedures

Exclusion criteria

• Allergy to fluorescein
• Colorectal cancer
• Advanced polyps or malignant lesions identified during colonoscopy
• Significant cardiopulmonary disease
• Renal failure
• Active infection or sepsis
• Severe IBD flare
• Use of medications that may affect gut permeability, including NSAIDs or antibiotics
• Inability to tolerate bowel preparation or safely complete the protocol
• Pregnancy or breastfeeding
• Inability or unwillingness to provide informed consent or comply with study procedures

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: No

Study design

Observational model

Cohort

Study locations

Center list to be confirmed — check the primary protocol.

Identifiers

NCT: NCT07546955 · REB 19656

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗