Neoplasms

All Conditions

Symptoms and General Pathology

Adenoma

Colonic Polyps

Hyperplasia

Polyps

Pathological Conditions, Anatomical

Neoplasms, Glandular and Epithelial

Neoplasms by Histologic Type

Herbal and Botanical

All Drugs and Chemicals

Cola

COLONIC POLYP HISTOLOGY PREDICTION IN WHITE LIGHT IMAGES COMBINING ARTIFICIAL INTELLIGENCE AND CLINICAL INFORMATION

AUTOMATED POLYP CLASSIFICATION

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.

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).

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

The results of the computer-aided system prediction will be compared with the final pathology report, which is the gold standard

Instituto de Salud Carlos III

Hospital Clinic of Barcelona

Principal Investigator

Polyps were identified by endoscopists assisted by an AI system: rectangular box marks; Monitoring of ileocecal position: whether blindness was reached was displayed in the lower left corner of the interface.

Mirror entry and exit time monitoring: the operation time is displayed in the upper left corner of the interface.

Colonoscopy withdrawal speed monitoring: the relative withdrawal speed was displayed on the left side of the interface.

ANDOANGEL

Honggang Yu, Doctor

In recent years, computer-aided diagnosis system based on artificial intelligence (AI) has been used in colorectal polyp detection. In recent years, computer-aided diagnosis system based on artificial intelligence (AI) has been used in colorectal polyp detection. However, whether AI-assisted can improve the adenoma-detection rate (ADR) is inconclusive. This study aims to evaluate the real-world performance of an AI system that combines polyp detection with colonoscopy quality control.

This study aims to explore the clinical application value of AI-based polyp detection and quality control function by comparing the data of polyp detection rate and adenoma detection rate in multiple centers with and without AI-assisted colonoscopy in a multicenter, prospective real world study. However, whether AI-assisted can improve the adenoma-detection rate (ADR) is inconclusive. This study aims to evaluate the real-world performance of an AI system that combines polyp detection with colonoscopy quality control.

This study aims to explore the clinical application value of AI-based polyp detection and quality control function by comparing the data of polyp detection rate and adenoma detection rate in multiple centers with and without AI-assisted colonoscopy in a multicenter, prospective real world study.

Artificial Intelligence-assisted System in Colonoscopy

To Evaluate the Effectiveness and Safety of an Artificial Intelligence-assisted System in Colonoscopy in a Real-world Obsevational Multicenter Study

The numerator is the number of patients who had at least one adenoma on colonoscopy, and the denominator is the total number of patients who underwent colonoscopy

The numerator is the number of patients who had at least one polyp on colonoscopy, and the denominator is the total number of patients who underwent colonoscopy

The numerator is the number of patients who had at least one serrated adenoma on colonoscopy, and the denominator is the total number of patients who underwent colonoscopy

The numerator is the total number of polyps detected by colonoscopy, and the denominator is the total number of colonoscopies

The pure negative time of the whole examination did not include the time of endoscopy entry and the time of biopsy

Renmin Hospital of Wuhan University

Colonoscope withdrawal was performed in the right colon evaluating the mucosa using standard white light.

Colonoscope withdrawal was performed in the right colon evaluating the mucosa using LCI (Linked Color Imaging).

Colonoscope withdrawal was performed in the right colon evaluating the mucosa using LCI (Linked color imaging)

LCI (Linked color imaging)

Franco Radaelli, MD

Although colonoscopy with polypectomy can prevent up to 80% of colorectal cancers, a significant adenoma miss rate still exists, particularly in the right colon. Optimizing the detection of adenomas and sessile serrated lesions in the right colon is crucial to increase the effectiveness of colonoscopy in colorectal cancer prevention.

Last generation Fuji videocolonscopes incorporates the Linked Color Imaging (LCI), a recently developed technology that differentiates the red colour spectrum more effectively than White Light imaging thanks to its optimal pre-process composition of light spectrum and advanced signal processing. The increased colour contrast results in more accurate delineation of abnormal inflammatory or neoplastic findings of colonic mucosa. Preliminary data suggest that LCI may be improve the detection of neoplastic lesion of colon.

The investigators performe a tandem prospective study to compare the right colon adenoma miss rates of LCI colonoscopy with those of conventional white light colonoscopy.

Therefore participants scheduled for colonoscopy for the assessment of symptoms or for colorectal cancer screening/surveillance receive the examination of the right colon twice, in a back to back fashion, with standard white light (WL) and with LCI. Patients are randomly assigned (1:1), via computer-generated randomisation with block size of 20, to which procedure is done first. The endoscopist are masked to group allocation until immediately before the cecum is reached. Examinations are performed with Fuji videocolonscopes series 700 (EC-760R, EC-760ZP).

Participants scheduled for colonoscopy for the assessment of symptoms or for colorectal cancer screening/surveillance receive the examination of the right colon twice, in a back-to-back fashion, with standard white light (WL) and with LCI (Linked Color Imaging). Patients are randomly assigned (1:1), via computer-generated randomisation with block size of 20, to which procedure is done first.

LCI (Linked Color Imaging) for Adenoma Detection in the Right Colon

Number of adenomas identified during the second right coloninspection/ overall number of adenomas identified during the first and the second right colon inspection

Number of advanced adenomas identified during the second right coloninspection/ overall number of advanced adenomas identified during the first and the second right colon inspection

Number of sessile serrated lesions identified during the second right coloninspection/ overall number of sessile serrated lesions identified during the first and the second right colon inspection

Study title	NCT ID	Conditions	Interventions	Eligibility	Organization
In Vivo Computer-aided Prediction of Polyp Histology on White Light Colonoscopy	NCT03775811	Colonoscopy Histology Computer-aided ... Artificial Inte... Colorectal Poly... Adenoma Colon P... Hyperplastic Po...	AUTOMATED POLYP...	18 Years -	Hospital Clinic of Barcelona
Artificial Intelligence-assisted System in Colonoscopy	NCT06406062	Adenoma Colon P...	ANDOANGEL	50 Years -	Renmin Hospital of Wuhan University
LCI (Linked Color Imaging) for Adenoma Detection in the Right Colon	NCT03279783	Adenoma Colon P...	LCI (Linked col...	18 Years - 80 Years	Valduce Hospital

Spots Global Cancer Trial Database for adenoma colon polyp