Artificial Intelligence and Dysplasia Detection in Inflammatory Bowel Disease (EIIDISIA Study)

Effect of an Artificial Intelligence System on the Detection of Dysplasia During Colonoscopy in Patients With Long-standing Inflammatory Bowel Disease: a Randomized Study (EIIDISIA Study)

Status

Active, not recruiting

Phases

Study type

Interventional

Source

ClinicalTrials.gov

Registry ID

NCT06281392

Enrollment

122

Registered

2024-02-28

Start date

2023-11-15

Completion date

2025-11-20

Last updated

2025-08-15

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Dysplasia, Inflammatory Bowel Diseases

Keywords

Inflammatory Bowel Disease, Virtual chromoendoscopy, NBI, CADe, Dysplasia

Brief summary

Randomized clinical study analyzing the efficacy of colonoscopy assisted by the Computer Aided Detection (CADe) system compared to virtual chromoendoscopy with Narrow Band Imaging (NBI) in the detection of colon dysplasia in patients with long-standing inflammatory bowel disease (IBD).

Interventions

DEVICENBI

Virtual colonoscopy assisted by NBI

DEVICECADe system

Colonoscopy assisted by an artificial intelligence system (CADe).

Study design

Allocation

RANDOMIZED

Intervention model

PARALLEL

Primary purpose

SCREENING

Masking

TRIPLE (Subject, Investigator, Outcomes Assessor)

Eligibility

Sex/Gender

ALL

Age

18 Years to No maximum

Healthy volunteers

Inclusion criteria

* Adults (age ≥18 years). * Patients with IBD meeting the following criteria: * Diagnosis of IBD confirmed by clinical and histological evidence. * Disease extent (Montreal classification): 1) For Ulcerative Colitis (UC), Extensive colitis (E3): involvement proximal to the splenic flexure; or, Left-sided colitis (E2): involvement distal to the splenic flexure. 2) For Crohn's Disease (CD): Colon-only involvement (L2); or, Ileocolonic involvement (L3). * Disease duration \> 7 years. * Patients with concommitant Primary Sclerosing Cholangitis, regardless of the extent of colonic involvement and disease duration. * Written consent from the patient to undergo colonoscopy (signed informed consent).

Exclusion criteria

* Personal history of colorectal cancer (CRC). * Previous colectomy (partial or complete). * Coagulopathy preventing biopsy or polypectomy/mucosal resection. * Colonoscopy performed in the last 6 months. * Pregnant or lactating women. * Ulcerative colitis with involvement limited to the rectum or Proctitis (Montreal E1). * Inadequate colonic preparation due to uncleavable fecal residue during colonoscopy: Boston Bowel Preparation Scale (BBPS) = 0 or 1 in any colon segment. * Presence of inflammatory activity due to IBD on endoscopy:1) For UC: Endoscopic Mayo Score ≥ 2. 2) For CD: Simplified Endoscopic Activity Score (SES-CD) \> 5. * Incomplete lower digestive endoscopy (failure to reach the cecum).

Design outcomes

Primary

Measure	Time frame	Description
Per lesion dysplasia detection rate (DDR)	Immediately after the procedure	Proportion of dysplastic lesions in relation to all the lesions analyzed that were detected by Artificial Intelligence System (CADe) or Virtual chromoendoscopy with NBI in patients with long-standing Inflammatory bowel disease who undergo surveillance colonoscopy
Per patient dysplasia detection rate (DDR)	Immediately after the procedure	Proportion of patients with dysplastic lesions in relation to all the patients with long-standing Inflammatory bowel disease included in the study detected by Artificial Intelligence System (CADe) or Virtual chromoendoscopy with NBI

Secondary

Measure	Time frame	Description
Duration of endoscopic examination	Immediately after the procedure	Median (IQR) examination time (in minutes) (including total, insertion, and withdrawal time) required by Artificial Intelligence System (CADe) or Virtual chromoendoscopy with NBI

Countries

Spain

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026