Artificial Intelligence-based Screening Models for Prevention and Early Detection of Colorectal Cancer

Trial statusRecruiting
Trial phaseNot applicable
Trial typeInterventional
Biological sexAll
Age50-69
SponsorIRCCS Azienda Ospedaliero-Universitaria di Bologna

About this trial

Colorectal cancer screening is based on the fecal occult blood test (FIT), which has low sensitivity for adenomatous polyps, based on the currently used cut-off. Risk factors such as obesity, diabetes, alcohol, and cigarette smoking are associated with the presence of high-risk neoplasia in the screening population. CADe systems appear to increase ADR in screening programs; however, uncertainty remains regarding their true effectiveness.

The study could provide the tools to:

1. devise a personalized pathway of CRC screening so as to refer to colonoscopy (with CAD or without CAD depending on the results that will be obtained) those at high risk of carrying neoplasms amenable to removal or curative treatment; 2. define risk categories for theoretical screening models giving the possibility of moving from the concept "one size fits all" to that of "personalized and precision prevention".

Eligibility criteria

This trial does not accept healthy volunteers

Qualifiers

Subjects aged 50-69 years who underwent the fecal occult blood screening program

Male and female sex

Positivity to fecal occult blood (FIT) on the immunochemical screening test (OC-Sensor);the cut-off for determination of test positivity is 20 micrograms HgB/gr of stool (cut-off used in our country).

Subjects who provided written consent to participate in the study.

Disqualifiers

Subjects who do not understand the Italian language

Subjects with previous history of total colectomy or bowel resections

Subjects with colonoscopy performed in the previous 12 months

Subjects with contraindications to colonoscopy

Trial design

Design model

Parallel

Treatments tested in this trial

  • GI Genius

    Device

    GI Genius software component is an artificial intelligence-based medical device that has been trained to process colonoscopy images containing regions consistent with colorectal lesions such as polyps, including those with flat morphology (nonpolyp).

  • Colonscope Olympus

    Other intervention

    Colonoscopy with one of the Olympus colonoscopes supplied to the IRCCS AOUBO Gastroenterology Unit without the aid of the CAD system

Treatment groups

1,400 Participants
are divided into 2 treatment groups
Group A: Colonoscopy using artificial intelligence (experimental group)Experimental treatment 1 intervention
Group B: Traditional colonoscopy (control group)Other 1 intervention

Trial outcomes

Primary outcomes

1

To develop a predictive risk model useful in tailoring screening strategies for high-risk colorectal neoplasm (advanced adenoma/early stage cancer)

Adenoma Detection Rate (ADR, percentage of patients screened with at least 1 adenoma or carcinoma histologically proven) assessed following endoscopic examination and post-collection histopathological result in relation to risk factors.

Time frame
From enrollment to the end of treatment at 18 months.

Secondary outcomes

Other outcomes

Sponsors and contacts

Click on the lead sponsor to view all of their trials.