Using AI-assisted Optical Polyp Diagnosis for Diminutive Colorectal Polyps (AI-OD)

Using Artificial Intelligence-assisted Optical Polyp Diagnosis for Diminutive Colorectal Polyps

This is a prospective study that is the first to implement resect and discard and diagnose and leave strategies in real-time practice using stringent documentation and adjudication by 2 expert endoscopists as the gold standard.

The primary aim of this study is to show the accuracy of intracolonoscopy AI-assisted optical diagnosis (CADx; autonomous or with human input) when the AI-assisted optical diagnosis made by the expert endoscopists is used as the reference standard. The specific aims are:

1. To evaluate the accuracy of intracolonoscopy AI-assisted optical polyp diagnosis (autonomous or with human input) by comparing it to the obtained optical histology diagnoses provided by two independent expert endoscopists as the reference standard.
2. To evaluate the agreement between the intracolonoscopy AI-assisted optical polyp diagnosis (autonomous or with human input) and the AI-assisted optical diagnosis performed by two independent expert endoscopists.
3. To determine whether AI-assisted optical polyp diagnosis for diminutive (1-5 mm) polyps can be implemented in routine clinical practice by demonstrating that at least 70% of the approached patients are interested in undergoing AI-assisted optical diagnosis (autonomous or with human input).
4. To evaluate the cost savings resulting from replacing pathology with AI-assisted optical diagnosis.

Study Overview

• Status: Recruiting
• Conditions: Artificial Intelligence
• Intervention / Treatment: Device: Artificial intelligence-assisted classification (CADx)

Detailed Description

All patients who meet the inclusion criteria can be enrolled. Eligible patients will be informed about the study through a consent form that includes information on optical diagnosis (resect and discard, diagnose and leave) and AI/CADx systems. We will ask the patients' willingness to undergo AI-assisted optical diagnosis with endoscopists' input (the first 102 patients) and automomous AI-assisted optical diagnosis (from patient 103 to 204). Subsequently, patients will be asked if they are willing to participate in the study, using AI-assisted optical diagnosis and the "resect and discard" and "diagnose and leave" strategy. If a patient declines to undergo optical diagnosis, they will be asked about the reason for their refusal to participate in the study. The options for their response include:

1. Concerns regarding undergoing an optical diagnosis.
2. Reluctance to participate in research projects in general.
3. Other reasons.
4. Preference not to answer the question.

Patients who agree to participate in the study will undergo standard colonoscopy procedures with AI-assisted optical diagnosis for all diminutive colorectal polyps identified. High-definition colonoscopes with a joint computer-assisted classification (CADx) support (CAD-EYE software EW10-EC02) will be used.

For the first 102 patients (i.e., the CAD-assisted optical diagnosis with endoscopist's input), the endoscopists will use the CAD-EYE blue light imaging (BLI) mode to enhance the visualization of polyp features. During the optical diagnosis using CADx, the most probable diagnosis (neoplastic or hyperplastic) will be displayed on the endoscopy screen. If the serrated pathology subtype is determined as the most probable histology, the endoscopists will make the final decision. They will also indicate whether their optical diagnosis was made with low or high confidence.

For the second group of 102 patients (i.e., autonomous CADx-assisted optical diagnosis), endoscopists will use CADx and BLI mode to perform optical diagnosis. Based on the CADx diagnosis, all 1-5 mm polyps diagnosed as hyperplastic or neoplastic will be resected and discarded, while those located in the rectosigmoid and diagnosed as hyperplastic will be left in the colon. When high-risk histology features are observed using BLI, in any patient, the endoscopists will inform the research assistant to document them, and the polyp will be sent for pathology examination in accordance with the ASGE PIVI guidelines recommendations. All polyps >5mm will be send for pathology evaluation. Polyp size will be measured using virtual scale technology integrated in the computer-assisted system (CAD) to ensure an accurate polyp sizing.18 A research assistant will document the characteristics of the detected polyps (i.e., location, size, morphology, AI-assisted intracolonoscopy and endoscopists optical diagnoses). All identified colorectal polyps will be removed following standard polypectomy practices. The entire colonoscopy procedures will be video recorded for quality assurance purposes. All diminutive polyps will be resected and discarded as part of the resect and discard strategy. Additionally, diminutive polyps located in the rectosigmoid colon will be detected and left in situ (diagnose and leave strategy) if no high-risk features are present.

• Study Type: Interventional
• Enrollment (Estimated): 204
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Canada
  • Quebec
    • Montréal, Quebec, Canada
      • Recruiting
      • Centre hospitalier de l'Université de Montréal
      • Contact: Julie Fleury; Phone Number: 30917 514-890-8000; Email: julie.fleury.chum@ssss.gouv.qc.ca
      • Contact: Daniel Von Renteln, MD

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Age 45-80 years
• Undergoing an outpatient colonoscopy at the Centre Hospitalier de l'Université de Montréal (CHUM)
• Signed informed consent form

Exclusion Criteria:

• Inflammatory Bowel Disease;
• Active colitis;
• Hereditary CRC syndrome;
• Coagulopathy;
• American Society of Anesthesiologists (ASA) status >3

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Diagnostic
• Allocation: Non-Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Other: AI-assisted classification with endoscopist's input; AI-assisted classification for diminutive polyps during a colonoscopy procedure using the CAD-eye detection and classification system, with input from the endoscopist in the case of serrated polyps, for patients who agree to undergo optical diagnosis of diminutive colorectal polyps.	Device: Artificial intelligence-assisted classification (CADx); CADeye (Fujifilm, Japan) is a joint detection (CADe) and classification (CADx) AI-supported system, which has been developed utilising AI deep learning technology to support endoscopic lesion detection and characterisation in the colon.; Other Names: CAD-eye
Other: Autonomous AI-assisted classification; AI-assisted classification for diminutive polyps during a colonoscopy procedure using the CAD-eye detection and classification system, with no input from the endoscopist, for patients who agree to undergo optical diagnosis of diminutive colorectal polyps.	Device: Artificial intelligence-assisted classification (CADx); CADeye (Fujifilm, Japan) is a joint detection (CADe) and classification (CADx) AI-supported system, which has been developed utilising AI deep learning technology to support endoscopic lesion detection and characterisation in the colon.; Other Names: CAD-eye

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy of the intracolonoscopy AI-assisted optical diagnosis	Using the optical diagnosis obtained from 2 experts as the reference, the accuracy of the intracolonoscopy AI-assisted optical diagnosis (OD, with endoscopist input or autonomous) is measured as the number of correct ODs out of all ODs (%).	120 days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Proportion of patients who accept study participation	Number of patients who accept to participate out of all patients approached	120 days
Proportion of the patients unwilling to participate due to concerns regarding undergoing an optical diagnosis	The proportion of the patients unwilling to participate in the study due to the concerns of resect and discard or diagnose and leave strategies (with endoscopist input or autonomous) and preference upon histopathological evaluation of all detected polyps	120 days
Agreement between the intracolonoscopy AI-assisted optical diagnosis and the AI-assisted optical diagnosis by experts	The agreement between AI-assisted OD (with and without endoscopist's input) and the experts' OD is measured as the number of cases where both ODs are the same out of all cases (%).	120 days
Diagnostic characteristics of AI-assisted optical diagnosis using adjudication by two expert endoscopists as the reference standard	To calculate the accuracy, sensitivity, specificity, positive predictive value, negative predictive value of (CADx) assisted optical diagnosis (resect and discard) strategy using adjudication by two expert endoscopists as the reference standard.	120 days
Proportion of polyps with a low-confidence diagnosis	Proportion of polyps with a low-confidence diagnosis intracolonoscopy and later by the expert independent reviewers	120 days
Assessing cost savings through AI-assisted optical diagnosis	The reduction in costs of the supplies and the pathology assessments according to the expenses reported by the CHUM gastroenterology and pathology departments including the expenses related to the supplies (e.g., snares, containers), immediate and delayed complications, specimen processing and equipment and physician fees.	120 days

Collaborators and Investigators

• Sponsor: Daniel Von Renteln
• Investigators: Principal Investigator: Daniel von Renteln, MD, Centre Hospitalier de l'Universite de Montreal (CHUM)

Study record dates

Study Major Dates

• Study Start (Actual): September 1, 2023
• Primary Completion (Estimated): May 1, 2025
• Study Completion (Estimated): June 30, 2025

Study Registration Dates

• First Submitted: September 22, 2023
• First Submitted That Met QC Criteria: September 22, 2023
• First Posted (Actual): September 28, 2023

Study Record Updates

• Last Update Posted (Actual): March 25, 2025
• Last Update Submitted That Met QC Criteria: February 14, 2025
• Last Verified: February 1, 2025

More Information

Terms related to this study

• Keywords: Colonoscopy; Optical diagnosis; Artificial intelligenc; Resect and Discard
• Additional Relevant MeSH Terms: Pathological Conditions, Anatomical; Polyps
• Other Study ID Numbers: 2024-11557/23.095

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No