What question did this study set out to answer?

To evaluate an AI-based tool for estimating colorectal polyp sizes during colonoscopy.

February 22, 2026Endoscopy0 citations

Author commentary on Giulio Antonelli et al.

Key Points

To evaluate an AI-based tool for estimating colorectal polyp sizes during colonoscopy.
Prospective multicenter study involving 225 patients and 465 polyps.
Evaluation of AI tool's accuracy for polyp size estimation during colonoscopy.
AI tool achieved an overall accuracy of 86% for polyp size estimation.
Accuracy improved to 93% specifically for diminutive polyps (≤5 mm).

Abstract

Giulio Antonelli et al. Clinical implications of computer-aided real-time size estimation of colorectal polyps during colonoscopy: a prospective study A new artificial intelligence (AI)-based tool for estimating lesion size was evaluated in this prospective multicenter study. Among 465 polyps (225 patients), the tool estimated colorectal polyp size with 86% accuracy overall, and 93% accuracy for diminutive polyps (≤5 mm). The authors suggest that this AI modality could standardize polyp size measurements during colonoscopy and facilitate adoption of cost-saving strategies based on optical diagnosis. Article published online: 20 February 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

Perguntar à IA

Bookmark

View Full Paper