NCT04693078

Detection of Colonic Polyps Via a Large Scale Artificial Intelligence (AI) System

Completed NA Results posted Last updated 3 March 2021

What this trial tests

NA trial testing AI polyp detection system based on deep learning in Colonic Polyp in 100 participants. Completed in 30 December 2020.

Timeline

18 May 2020

Primary endpoint
30 November 2020

30 December 2020

Quick facts

Lead sponsor	Shaare Zedek Medical Center
Phase	NA
Status	Completed
Study type	INTERVENTIONAL
Allocation	na
Design	single group
Masking	none
Primary purpose	screening
Enrollment	100
Start date	18 May 2020
Primary completion	30 November 2020
Estimated completion	30 December 2020
Sites	1 location across Israel

Drugs / interventions tested

• AI polyp detection system based on deep learning

Conditions studied

• Colonic Polyp — all drugs for Colonic Polyp →

Sponsor

Shaare Zedek Medical Center

Who can join

Adults 40 to 80, any sex, with Colonic Polyp. Patients with the condition only — healthy volunteers not accepted.

Results — posted to ClinicalTrials.gov

Per-arm endpoint measurements with 95% confidence intervals where reported. Source: trial results section.

Number of Additional Polyps Detected by the DEEP System in Real Time Colonoscopy Primary · Through study completion, an average of 12 months

During the colonoscopy procedure, in real time when a polyp is found, the colonoscopist will rate the polyp as an elusive polyp detected by the system that might have been missed or a polyp that would have been detected with or without the system. The outcome measure will be reported as the average of additional polyps detected per colonoscopy by the DEEP system

Group	Value	95% CI
Intervention Arm	0.89	0.66 – 1.12

The Rate of Adverse Events During the Study Attributed or Not to the Use of the DEEP System Primary · Until discharge, assessed up to 7 days

Prospective assessment adverse events during the study. The following adverse event will be monitored: Perforation, bleeding, and cardiorespiratory adverse events during the procedure

Group	Value	95% CI
Intervention Arm	0

Rate of False Positives (False Alarms) Per Colonoscopy Secondary · Through study completion, an average of 12 months

During the colonoscopy procedure, in real time after each polyp found by the DEEP system, the colonoscopist will rate the polyp as either a true polyp or a false positive detection or a "false alarm" this measure will be reported as the average of false positive detection per colonoscopy

Group	Value	95% CI
Intervention Arm	3.87	3.34 – 4.40

Colonoscopist User Experience While Using the DEEP System in a 5 Point Scale Secondary · Through study completion, an average of 12 months

At the end of the procedures the colonoscopist will be requires to answer the question "from a scale of 1-5 how useful did you find the system in this procedure?", where higher scores represent more usefulness. This measure will be reported as the average score form all 100 procedures.

Group	Value	95% CI
Intervention Arm	3.9	3.7 – 4.1

Sponsor's own description

Colonoscopy is the gold standard for detection and removal of precancerous lesions, and has been amply shown to reduce mortality. However, the miss rate for polyps during colonoscopies is 22-28%, while 20-24% of the missed lesions are histologically confirmed precancerous adenomas. To address this shortcoming, the investigators propose a new polyp detection system based on deep learning, which can alert the operator in real-time to the presence and location of polyps during a colonoscopy. The investigators dub the system DEEP: (DEEP) DEtection of Elusive Polyps. The DEEP system was trained on 3,611 hours of colonoscopy videos derived from two sources, and was validated on a set comprising 1,393 hours of video, coming from a third, unrelated source. For the validation set, the ground truth labelling was provided by offline gastroenterologist annotators, who were able to watch the video in slow-motion and pause/rewind as required; two or three specialist annotators examined each video.

This is a prospective, non-blinded, non-randomized pilot study of patients undergoing elective screening and surveillance colonoscopies using DEEP.

The aim of the study is to:

Assess the:

1. Number of additional polyps detected by the DEEP system in real time colonoscopy.
2. Safety by prospective assessment of the rate of adverse events during the study period attributed or not to the use of the DEEP system.
3. Stability of the DEEP system by measuring the rate of false positives (False Alarms) per colonoscopies 4 And to examine its feasibility and usefulness of in clinical practice by assessing the colonoscopist user experience while using the DEEP system in a 5 point scale.

Publications & conference data

3 peer-reviewed publications reference this trial (live from Europe PMC):

1. Detection of elusive polyps using a large-scale artificial intelligence system (with videos).
   Livovsky DM, Veikherman D, Golany T, Aides A, et al · · 2021 · cited 16× · PMID 34216598 · DOI 10.1016/j.gie.2021.06.021
2. Characteristics of Artificial Intelligence Clinical Trials in the Field of Healthcare: A Cross-Sectional Study on ClinicalTrials.gov.
   Wang A, Xiu X, Liu S, Qian Q, et al · · 2022 · cited 12× · PMID 36294269 · DOI 10.3390/ijerph192013691
3. Identification of technology frontiers of artificial intelligence-assisted pathology based on patent citation network.
   Zhang T, Chen J, Lu Y, Yang X, et al · · 2022 · cited 6× · PMID 35994484 · DOI 10.1371/journal.pone.0273355