Artificial intelligence based polyp size measurement in gastrointestinal endoscopy using the auxiliary water jet as a reference.

Background and study aims Measurement of colorectal polyp size during endoscopy is mainly performed visually. In this work, we propose a novel polyp size measurement system (Poseidon) based on artificial intelligence (AI) using the auxiliary water jet as a measurement reference. Methods Visual estimation, biopsy forceps-based estimation, and Poseidon were compared using a CT-colonography-based silicone model with 28 polyps of defined sizes. Four experienced gastroenterologists estimated polyp sizes visually and with biopsy forceps. Furthermore, the gastroenterologists recorded images of each polyp with the water jet in proximity for the application of Poseidon. Additionally, Poseidon's measurements of 29 colorectal polyps during clinical routine were compared to visual estimates. Results Visual estimation had the largest median percentage error (PE) of 25.2% (95% confidence interval (CI95%): 19.1, 30.4), followed by biopsy forceps-based estimation with median 20% (14.4, 25.6) in the silicone model. Poseidon presented a significantly lower median PE of 7.4% (5, 9.4; p<0.001) than other methods. During routine colonoscopies, Poseidon presented a significantly lower median PE (7.7% 6.1, 9.3) than visual estimation (22.1% 15.1, 26.9; p<0.001). Conclusion In this work, we present a novel AI-based method for measuring colorectal polyp size with significantly higher accuracy than other common sizing methods.