A17 COMPARING ENDOSCOPIST DIAGNOSIS OF COLORECTAL POLYPS ASSISTED BY ARTIFICIAL INTELLIGENCE (CADX) VS CADX WITHOUT ENDOSCOPIST INPUT: A RANDOMIZED CONTROLLED TRIAL

TL;DR

The study compares AI-assisted colonoscopy diagnosis with and without endoscopist input, finding similar accuracy in both approaches.

Contribution

This is the first randomized trial comparing AI-only diagnosis with endoscopist-assisted AI diagnosis for colorectal polyps.

Findings

CADx without human input had 76.3% accuracy, similar to 70.5% in the endoscopist-assisted group.

Sensitivity and specificity for adenoma detection were comparable between the two groups.

Results suggest AI alone could support resect-and-discard strategies without endoscopist involvement.

Abstract

Artificial intelligence-based optical diagnosis systems (CADx) have been developped to assist in eliminating the need for histologic diagnosis of diminutive colorectal polyps (resect-and-discard and diagnose-and-leave strategies). However, these systems have not yet been implemented in routine clinical practice.

We were interested in evaluating the performance of CADx without human input to diagnoses performed by endoscopists assisted by CADx.

We performed a randomized clinical trial of patients undergoing elective colonoscopies at the CHUM. Patients were randomized into two arms: 1) optical diagnosis of colorectal polyps using CADx without human input; 2) endoscopists performed optical diagnosis after consulting a real-time CADx diagnosis (Human in the Loop [HiL]). Primary outcome was accuracy in optical diagnosis for both arms.

467 patients were randomized (229 in the CADx group, 238 in the HiL group). Overall accuracy for optical diagnosis was 76.3% in the CADx group and 70.5% in the HiL group (p=0.19). Sensitivity, specificity, PPV and NPV for adenoma diagnosis were 89.7%, 58.4%, 82.2%, and 72.5% respectively in the CADx group vs 85.3%, 69.6%, 81.8%, and 74.8% in the HiL group. Sensitivity, specificity, PPV and NPV did not differ significantly between the two groups.

Optical diagnosis of colorectal polyps had similar accuracy when using CADx without human input compared to endoscopist-based diagnosis assisted by CADx. Resect and discard and diagnose and leave strategies can therefore potentially be implemented without need for endoscopist optical diagnosis.

CAGFujifilm