Fig. 1: CONSORT recruitment flowchart.

Seventeen patients were withdrawn before the colonoscopy, 12 of which were not randomized. Thirty-four patients were withdrawn during the procedure (20 procedures abandoned (16 due to inadequate bowel prep preventing intubation, one due to stenosis, one due to nausea, one due to hypoxia and one unknown), five identified during the procedure to not meet patients eligibility criteria, four procedures used unapproved endoscopes per-protocol, three technical issues (e.g. computer not connected to the endoscopy stack) and two patients withdrawals). Twenty-one completed procedures were excluded after it was identified post-procedurally to be a patient with positive fecal immunochemical test (FIT) or the procedure used CF-EZ1500DI endoscopes which were not cleared for use in the US by the end of the study. A further 72 were excluded. This included 25 procedures from endoscopists who failed to complete the minimum ten procedures required, 19 with inadequate bowel preparation (BBPS < 2 in any one of the three colonic segments), 14 with a withdrawal time below 6 min, ten due to the use of unapproved equipment (six unapproved scopes per-protocol, two performed withdrawal in TXI or NBI, two procedures used a different CADe system from ‘CADDIE’), two randomization errors where the procedure was not performed according to the randomized allocation, and two patients only identified at the end of the procedure to not meet patients eligibility criteria. Image created using Microsoft PowerPoint, version 16.103.2 (2025).