Accuracy of CADx System for White Light Colonic Polyp Characterization (GIC)

Accuracy and Feasibility of CADx System for White Light Colonic Polyp Characterization

The endoscopist performances in the optical diagnosis (OD) of colonic polyps with the available technologies vary widely across centers and across endoscopists. The OD process is strictly related to the operator training and expertise. Most of the available studies in optical characterization have been carried out by expert endoscopist in tertiary high volume centers, and weren't replied on large unselected populations. For these reasons, at the moment the optical characterization of polypoid lesions can't replace, in the everyday clinical practice, the histopathological evaluation of resected polyps. Artificial intelligence (AI)-based systems have the potential to make optical characterization process of colonic polyps easier and more reliable, thus supporting the endoscopist in the application of leave-in-situ and of resect-and-discard strategies. The implementation of such strategies would lead to a significant economic saving and a decrease of risks and complications related to unnecessary polypectomy. GI-Genius System (Medtronic Inc, Minneaopolis, USA) is a CNN-based algorithm allowing an automatic OD of colonic polyps. This system does not require dedicated light setting for polyp evaluation as it works with white light high definition images, which are the actual standard in every endoscopic unit. During colonoscopy, when a polyp is framed within the screen, a green detection box surrounds the polyp and the system automatically provides (whenever possible) the optical diagnosis labeling the polyp as "adenoma or non-adenoma". When the automatic polyp charaterization is unfeasible the label "no prediction" appears. Nowadays only few data about the feasibility and performances of this system in clinical practice are available. In addition published studies are mostly focused on technical rather thann clinical issues. The present prospective observational trial is primarily aimed at evaluating the diagnostic accuracy of optical characterization of colonic polyps <= 1 cm using GI-Genius System in daily clinical practice, having histopathology examination as reference standard.

Study Overview

• Status: Completed
• Conditions: Colonic Polyp; Adenoma Colon

• Study Type: Observational
• Enrollment (Actual): 500

Contacts and Locations

• Study Contact: Name: Franco Radaelli, MD; Phone Number: 0039031324147; Email: francoradaelli01@gmail.com
• Study Contact Backup: Name: Emanuele Rondonotti, MD; Phone Number: 0039031324147; Email: ema.rondo@gmail.com

Study Locations

• Italy
  • Como, Italy, 22100
    • Gastroenterology Unit, Valduce Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 85 years (Adult, Older Adult)
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Consecutive adults (18-85 years) outpatients referred for colonoscopy are invited to participate in the present study.

Description

Inclusion Criteria:

Consecutive outpatients undergoing colonoscopy for one of the following indication:

• FIT-based screening program
• voluntary screening
• post-polipectomy surveillance
• gastrointestinal symptoms
• family history of colorectal cancer.

Exclusion Criteria:

• urgent colonoscopy
• patients in which polyps could not be resected (e.g ongoing anticoagulation therapy)
• patients with past colorectal cancer or polyposis syndrome

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic accuracy of CADx in evaluating small colonic polyps	Numer of correctly characterized polyps using AI/total number of detected polyps	Entire study duration (3 months)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Feasibility of CADx in evaluating small colonic polyps	Number of characterizable polyps using AI/total number of detected polyps	Entire study duration (3 months)
Sensitivity	CADx sensitivity (rate of TP/TP+FN) will be calculated having histopathology as reference standard and according to polyp size and location	Entire study duration (3 months)
specificity	CADx specificity (rate of TN/TN+FP) will be calculated having histopathology as reference standard according to polyp size and location	Entire study duration (3 months)
Positive predictive value	CADx positive predictive value (rate of TP/TP+FP) will be calculated having histopathology as reference standard according to polyp size and location	Entire study duration (3 months)
Negative predictive value	CADX negative predictive value (rate of TN/TN+FN) will be calculated having histopathology as reference standard according to polyp size and location	Entire study duration (3 months)

Collaborators and Investigators

• Sponsor: Valduce Hospital

Study record dates

Study Major Dates

• Study Start (Actual): August 5, 2022
• Primary Completion (Actual): November 5, 2022
• Study Completion (Actual): November 5, 2022

Study Registration Dates

• First Submitted: August 5, 2022
• First Submitted That Met QC Criteria: August 5, 2022
• First Posted (Actual): August 8, 2022

Study Record Updates

• Last Update Posted (Actual): March 26, 2024
• Last Update Submitted That Met QC Criteria: March 24, 2024
• Last Verified: March 1, 2024

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Neoplasms by Histologic Type; Neoplasms; Neoplasms, Glandular and Epithelial; Pathological Conditions, Anatomical; Intestinal Polyps; Adenoma; Polyps; Colonic Polyps
• Other Study ID Numbers: 41/2022

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No