CADDIE Trial - Computer Aided Diagnosis and Detection for Intelligent Endoscopy (CADDIE)

Multi-Centre, Open-label, Randomised, Prospective Trial to Assess Efficacy and Safety of the CADDIE Artificial Intelligence System for Improving Endoscopic Detection of Colonic Polyps in Real-time.

Background:

Colonoscopy is accepted to be the gold standard for screening of colorectal cancer (CRC). Most CRCs develop from adenomatous polyps, with colonoscopy accepted to be the gold standard for screening of CRC. An endoscopist's ability to detect polyps is assessed in the form of an Adenoma Detection Rate (ADR). Each 1.0% increase in ADR is associated with a 3.0% decrease in the risk of the patient developing an interval CRC. There remains a wide variation in endoscopist ADR.

More recently, the use of artificial intelligence (AI) and computer aided diagnosis in endoscopy has been gaining increasing attention for its role in automated lesion detection and characterisation. AI can potentially improve ADR, but previous AI related work has largely focused on retrospectively assessing still endoscopic images and selected video sequences which may be subject to bias and lack clinical utility. There are only limited clinical studies evaluating the effect of AI in improving ADR.

The CADDIE device uses convolutional neural networks developed for computer assisted detection and computer assisted diagnosis of polyps.

Primary objective: To determine whether the CADDIE artificial intelligence system improves endoscopic detection of adenomas during colonoscopy.

Primary endpoint: The difference in adenoma detection rate (ADR) between the intervention (supported with the CADDIE system) and non-intervention arm

Study design: Multi-Centre, open-label, randomised, prospective trial to assess efficacy and safety of the CADDIE artificial intelligence system for improving endoscopic detection of colonic polyps in real-time.

Study Overview

• Status: Completed
• Conditions: Polyps; Colon Polyp; Adenoma Colon; Colorectal Polyp
• Intervention / Treatment: Device: CADDIE- Computer Aided (AI) Device used in Endoscopy

• Study Type: Observational
• Enrollment (Actual): 739

Contacts and Locations

Study Locations

• United Kingdom
  • London, United Kingdom
    • University College London

Participation Criteria

Eligibility Criteria

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

• Sampling Method: Non-Probability Sample

Study Population

Study population will include patient's scheduled to undergo a surveillance or symptomatic colonoscopy at the participating sites.

Description

Inclusion Criteria:

• Patient's scheduled to undergo a surveillance or symptomatic colonoscopy with an endoscopist participating in the study phase of the trial.
• Male and female participants aged 18 years or older at the time of informed consent.
• Patient's able to comprehend, sign and date the written informed consent document to participate in the study.

Exclusion Criteria:

• Emergency colonoscopies and/or inpatient colonoscopies.
• Patients with inflammatory bowel disease (IBD)
• Patients with Colorectal Cancer (CRC)
• Patients with previous CRC
• Patients with previous colonic resection
• Patients returning for a planned elective therapeutic colonoscopy.
• Polyposis syndromes
• Current or relevant history of a physical or psychiatric illness or any medical condition that in the opinion of the investigator could affect the patient's safety or interfere with the study assessments

• Patients with a contraindication for biopsy or polypectomy. These include:

  • Patients who have not withheld medications pre-disposing to bleeding at time of colonoscopy as per local site /national guidelines 30.
  • Patients with a history of haemostasis disorders (Haemostasis disorders will include but will not be limited to: patients with haemophilia or other congenitally acquired clotting factor deficiencies, patients with cirrhosis with coagulopathy, patients known to have thrombocytopenia (<80,000 platelet/ul) and individuals with Von Willebrand's disease or other known platelet malfunction disorders).
• Patients is enrolled in another research study with an investigational medicinal product (IMP) or non-IMP that pre-disposes them to bleeding

Study Plan

How is the study designed?

Design Details

• Observational Models: Cohort
• Time Perspectives: Prospective

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
CADDIE; The endoscopist will be assisted with CADDIE system to detect polyps. The endoscopist will perform optical diagnosis of polyps with the assistance of the CADDIE's polyp characterisation function.	Device: CADDIE- Computer Aided (AI) Device used in Endoscopy; CADDIE assisted polyp detection; CADDIE assisted polyp characterisation
Standard Procedure; In addition to routine colonoscopy the endoscopist will perform optical diagnosis of detected polyps without the assistance of the CADDIE.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
To Evaluate Whether Using the CADDIE System Improves Endoscopist Detection of Adenomas During Colonoscopy.	The difference in endoscopist ADR between the intervention (CADDIE system) and non-intervention arm.	30 days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Evaluate the Difference in Number of Adenomas Detected Per Colonoscopy Between Intervention and Non-intervention Arm	Compare the difference in the number of adenomas detected per colonoscopy between the intervention and non-intervention arm	30 days
To Determine Whether the CADDIE Artificial Intelligence System Improves Endoscopic Detection of All Polyps During Colonoscopy	Polyp Detection Rate (including proximal polyp detection rate) in the interventional group compared to the control group. - Mean number of polyps detected per colonoscopy in the interventional group compared to the control group.	30 days
Compare the Accuracy of Endoscopist Optical Diagnosis of Diminutive Polyps When Using the CADDIE System (Intervention Arm) Compared to Without the CADDIE System (Control Arm)	Compare the accuracy of CADDIE against human endoscopist (high confidence diagnoses) for the optical diagnosis of diminutive polyps. Accuracy is defined as true positive and true negative divided by diagnoses total polyps that were optically diagnosed with high confidence.	30 days
Compare the Accuracy of Endoscopist Optical Diagnosis to Assign Colonoscopy Surveillance Intervals When Using the CADDIE System (Intervention Arm) Compared to Without the CADDIE System (Control Arm)	Compare the accuracy of using CADDIE's optical diagnosis to assign surveillance colonoscopy intervals against using human endoscopist optical diagnosis (high confidence diagnoses) for assigning surveillance colonoscopy intervals.	30 days
Compare the Accuracy of Endoscopist Optical Diagnosis of Diminutive Rectal Polyps When Using the CADDIE System (Intervention Arm) Compared to Without the CADDIE System (Control Arm)	Compare the accuracy of CADDIE against human endoscopist (high confidence diagnoses) for the optical diagnosis of diminutive rectal polyps.	30 days
Evaluate the Safety of the CADDIE by Recording Occurrence and Severity of Adverse Events.	Safety of the device will be assessed through monitoring of adverse events for 30 days' post-procedure. Adverse events are defined as: Abdominal pain/discomfort.; Abdominal bloating.; Sedation related aspiration.; Bleeding risk.; Perforation.; Post polypectomy electrocoagulation syndrome.; Infection.	30 days
Assess Patient Anxiety During the Colonoscopy	Patient response to the question 'How anxious or calm were you during the colonoscopy?' assessed by Likert Scale with 5 options - 1 (Very anxious), 2 (Somewhat anxious), 3 (Neither Anxious nor calm), 4 (Somewhat calm), 5 (Very calm).	30 days

Collaborators and Investigators

• Sponsor: University College, London
• Investigators: Principal Investigator: Manish Chand, MBBS FRCS PhD, UCL

Study record dates

Study Major Dates

• Study Start (Actual): April 29, 2021
• Primary Completion (Actual): March 8, 2023
• Study Completion (Actual): March 8, 2023

Study Registration Dates

• First Submitted: March 11, 2020
• First Submitted That Met QC Criteria: March 25, 2020
• First Posted (Actual): March 30, 2020

Study Record Updates

• Last Update Posted (Actual): March 25, 2025
• Last Update Submitted That Met QC Criteria: February 18, 2025
• Last Verified: February 1, 2025

More Information

Terms related to this study

• Keywords: Endoscopy; Artificial Intelligence; ADR
• Additional Relevant MeSH Terms: Neoplasms; Pathological Conditions, Anatomical; Neoplasms by Histologic Type; Neoplasms, Glandular and Epithelial; Adenoma; Polyps
• Other Study ID Numbers: 126166

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Patients will be asked informed consent with the possibility to share the data
• IPD Sharing Time Frame: Data will be made available from 1 year after the trial findings are published. This is available for up to 5 years after study completion.
• IPD Sharing Access Criteria: We will make provision for and consider data sharing requests from bona fide researchers who must abide by the following principles: data will be collected with an high level of quality assurance, data will be held securely with appropriate documentation, data will not be put into the public domain or otherwise shared without explicit ethical review or legal obligation, and they will aim to use any data generated to the maximum public good. These datasets are governed by data usage policies specified by the data controller (UCL). If a conflict exists that severely restricts the analyses that can be undertaken, we would endeavour to support outside researchers by hosting them as visiting workers in our team so that they can access the data. We are committed to complying with the Wellcome Trust Data Sharing Policy. Applications are subject to review by Prof Lovat and Rawen Kader and should be emailed to l.lovat@ucl.ac.uk

Drug and device information, study documents

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