Deep learning technology has an increasing role in medical image applications and, recently, an artificial intelligence device has been developed and commercialized by Medtronic for identification of polyps during colonoscopy (GI-GENIUS). This kind of computer-aided detection (CADe) devices have demonstrated its ability for improving polyp detection rate (PDR) and the adenoma detection rate (ADR). However, this increase in PDR and ADR is mainly made at the expense of small polyps and non advanced adenomas. Colonoscopies after a positive fecal immunochemical test (FIT) could be the scenario with a higher prevalence of advanced lesions which could be the ideal situation for demonstrating if these CADe systems are able also to increase the detection of advanced lesions and which kind of advanced lesions are these systems able to detect. The CADILLAC study will randomize individuals within the population-based Spanish colorectal cancer screening program to receive a colonoscopy where the endoscopist is assisted by the GI-GENIUS device or to receive a standard colonoscopy. If our results are positive, that could suppose a big step forward for CADe devices, in terms of definitive demonstration of being of help for efectively identify also advanced lesions.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
DIAGNOSTIC
Masking
NONE
Enrollment
3,400
Colonoscopy assisted by GI-GENIUS device
Standard colonoscopy
Hospital General Universitario de Alicante
Alicante, Spain
Hospital Clinic Barcelona
Barcelona, Spain
Complexo Hospitalario de Ourense
Ourense, Spain
Hospital Universitario Central de Asturias
Oviedo, Spain
Hospital Universitario Río Hortega
Valladolid, Spain
Hospital Universitario Álvaro Cunqueiro
Vigo, Spain
Detection of advanced lesions
To determine the impact of the GI-GENIUS device as an assistant to the endoscopist to detect advanced lesions (advanced adenomas and advanced serrated polyps) in FIT-based screening colonoscopies.
Time frame: 15 months
Detection of other type of lesions
To determine the ability of the GI-GENIUS device to detect other lesions (colorectal cancer, adenomas, advanced adenomas, serrated polyps, advanced serrated polyps, non-polypoid lesions ≥10mm and right-sided advanced adenomas)
Time frame: 15 months
Characteristics of GI-GENIUS
To describe the characteristics of using the GI-GENIUS device during a colonoscopy: withdrawal time, false-positive and false-negative rate of advanced lesions using the device, etc.
Time frame: 1 month
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.