Artificial Intelligence, CADe, Polyps of Colon, Surveillance Endoscopy for Colorectal Cancer
Conditions
Brief summary
Evaluation of an artificial intelligence system for polyp detection (CADe)
Detailed description
We aim to evaluate the artificial intelligence system ENDOMIND that supports endoscopist in detection of polyps during surveillance endoscopy for colorectal cancer. 1070 patients out of 6 gastroenterologic practice are randomized 1:1 for conventional surveillance colonoscopy vs. surveillance with AI support. Primary endpoint is Adenoma detection rate.
Interventions
DEVICEENDOMIND
CADe system for Polyp detection
Sponsors
Klinikum Stuttgart
Wuerzburg University Hospital
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* indication CRC surveillance endoscopy * indication post-polypectomy surveillance endoscopy * positive fecal immunochemical test in patients \>=50 years
Exclusion criteria
* reasonable suspicion of inflammatory bowl disease * reasonable suspicion of familiar polyposis Syndrome * Patient after radiation/resection of colonic parts
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| adenoma detection rate | 4 months | proportion of individuals undergoing a complete screening colonoscopy who have one or more adenomas |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| polyp detection rate | 4 months | Exmainations with minimum one polyp detected. |
| withdrawal time | 4 months | Time of withdrawal. |
| resection time | 4 months | Time spent on polyp resections. |
| Boston Bowl Preparation Score | 4 months | minimum 0, Maximum 9; should be higher than 5 for appropriate surveillance |
Countries
Germany
Outcome results
None listed