In Vivo Computer-aided Prediction of Polyp Histology on White Light Colonoscopy

Our group, prior to the present study, developed a handcrafted predictive model based on the extraction of surface patterns (textons) with a diagnostic accuracy of over 90%24. This method was validated in a small dataset containing only high-quality images.

Artificial intelligence is expected to improve the accuracy of colorectal polyp optical diagnosis. We propose a hybrid approach combining a Deep learning (DL) system with polyp features indicated by clinicians (HybridAI). A pilot in vivo experiment will carried out.

Study Overview

• Status: Completed
• Conditions: Artificial Intelligence; Colonoscopy; Colorectal Polyp; Histology; Hyperplastic Polyp; Computer-aided Diagnosis; Adenoma Colon Polyp
• Intervention / Treatment: Other: AUTOMATED POLYP CLASSIFICATION

Detailed Description

Optical diagnosis aims to predict the histology of a polyp based on its endoscopic features. This practice could avoid histopathological analysis and reduce the derived costs. Under this premise, the American Society of Gastrointestinal Endoscopy (ASGE), in its Preservation and Incorporation of Valuable endoscopic Innovations (PIVI) statement, established a diagnostic threshold for real-time endoscopic assessment of diminutive polyps. The rationale for its implementation is that the prevalence of advanced histology in polyps < 5mm is very low (0.5%).

Several studies have demonstrated that optical diagnosis of small polyps is safe and feasible in clinical practice and comparable to the current gold standard, histopathology. However, the accuracy of optical diagnosis has been shown to be insufficient in community-based practices or in non-expert hands and the diagnosis is even more difficult in diminutive polyps < 3 mm in which the discrepancy between the endoscopic and pathological diagnosis is about 15%.

Artificial Intelligence (AI) has emerged as a help tool for polyp characterization.

Aiming to improve optical diagnosis using AI methods, we propose a hybrid approach that combines DL with characteristics of polyps manually indicated by endoscopists (HybridAI).

• Study Type: Observational
• Enrollment (Actual): 90

Contacts and Locations

Study Locations

• Spain
  • Barcelona, Spain, 08036
    • Hospital Clinic de Barcelona

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

All patients with polyps of any size/morphology, detected in a routine or screening colonoscopy, that are resected endoscopically and recovered for histological analysis will be included.

The images obtained will be used to expand the database.

Description

Inclusion Criteria:

• Age > 18 years
• Approval of participation in the study. Signature of informed consent
• Patients with at least one polyp of any size/morphology diagnosed in a routine or screening colonoscopy
• Endoscopies performed with high definition endoscopes

Exclusion Criteria:

• Age <18 years
• Refusal to participate in the study
• Polyps partially resected in a previous endoscopy
• Patients with inflammatory disease
• Impossibility to wash remains of stool or mucus on the surface of the polyp

Study Plan

How is the study designed?

Design Details

• Observational Models: Cohort
• Time Perspectives: Prospective

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy of the computer-aided system for predicting polyps histology in real clinical practice	The results of the computer-aided system prediction will be compared with the final pathology report, which is the gold standard	One year

Collaborators and Investigators

• Sponsor: Hospital Clinic of Barcelona
• Collaborators: Instituto de Salud Carlos III

Publications and helpful links

General Publications

• Byrne MF, Chapados N, Soudan F, Oertel C, Linares Perez M, Kelly R, Iqbal N, Chandelier F, Rex DK. Real-time differentiation of adenomatous and hyperplastic diminutive colorectal polyps during analysis of unaltered videos of standard colonoscopy using a deep learning model. Gut. 2019 Jan;68(1):94-100. doi: 10.1136/gutjnl-2017-314547. Epub 2017 Oct 24.
• Sanchez-Montes C, Sanchez FJ, Bernal J, Cordova H, Lopez-Ceron M, Cuatrecasas M, Rodriguez de Miguel C, Garcia-Rodriguez A, Garces-Duran R, Pellise M, Llach J, Fernandez-Esparrach G. Computer-aided prediction of polyp histology on white light colonoscopy using surface pattern analysis. Endoscopy. 2019 Mar;51(3):261-265. doi: 10.1055/a-0732-5250. Epub 2018 Oct 25.
• Bernal J, Histace A, Masana M, Angermann Q, Sanchez-Montes C, Rodriguez de Miguel C, Hammami M, Garcia-Rodriguez A, Cordova H, Romain O, Fernandez-Esparrach G, Dray X, Sanchez FJ. GTCreator: a flexible annotation tool for image-based datasets. Int J Comput Assist Radiol Surg. 2019 Feb;14(2):191-201. doi: 10.1007/s11548-018-1864-x. Epub 2018 Sep 25.

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2019
• Primary Completion (Actual): March 31, 2019
• Study Completion (Actual): December 31, 2022

Study Registration Dates

• First Submitted: December 11, 2018
• First Submitted That Met QC Criteria: December 11, 2018
• First Posted (Actual): December 14, 2018

Study Record Updates

• Last Update Posted (Actual): January 18, 2023
• Last Update Submitted That Met QC Criteria: January 17, 2023
• Last Verified: January 1, 2023

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Neoplasms by Histologic Type; Neoplasms; Neoplasms, Glandular and Epithelial; Pathological Conditions, Anatomical; Adenoma; Polyps
• Other Study ID Numbers: HISINVIA; PI17/00894 (Other Grant/Funding Number: Instituto de Salud Carlos III)

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