用于确定结肠镜检查中人工智能检测部位的内镜医师识别率的眼动追踪分析

Eye Tracking Analysis to Determine the Endoscopist's Recognition Rate for Artificial Intelligence-Detected Sites in Colonoscopy.

作者信息

Ishibashi Fumiaki, Suzuki Sho, Mochida Kentaro, Nagai Mizuki, Ozaki Eri, Okusa Kosuke

机构信息

Department of Gastroenterology, International University of Health and Welfare Ichikawa Hospital, 6-1-14 Konodai, Ichikawa-Shi, Chiba, 272-0827, Japan.

Graduate School of Medicine, International University of Health and Welfare, Tokyo, 107-8402, Japan.

出版信息

Dig Dis Sci. 2025 Jul 24. doi: 10.1007/s10620-025-09205-6.

DOI:10.1007/s10620-025-09205-6

PMID:40707744

Abstract

PURPOSE

Computer-aided detection systems (CADe) are used in screening colonoscopy; however, no study has quantitatively analyzed how endoscopists recognize candidate sites identified by CADe. The aim of this study was to determine the extent to which endoscopists could recognize CADe-detected sites.

METHODS

This prospective study used a dedicated eye tracking technology-based system to record the endoscopist's eye position and CADe detection sites in a single video during colonoscope withdrawal. All videos were reviewed to calculate the endoscopist's recognition rate, which was defined as the percentage of sites identified by CADe at which the endoscopist stared for > 300 ms. A logistic regression model was used to determine the predictors for quicker lesion detection by endoscopists than by CADe.

RESULTS

In 55 colonoscopies, 447 sites (306 artifacts, 98 adenomas, 28 hyperplastic polyps, 12 sessile serrated lesions [SSLs], 3 intramucosal cancers) were detected by CADe. The overall endoscopist recognition rate was 82.1%; the rate was 95.9% for adenomas, 92.8% for hyperplastic polyps, 100% for SSLs, 100% for intramucosal cancers, and 75.8% for artifacts. However, tracking speed analysis showed that only 33.3% CADe-detected SSLs were tracked by endoscopists within 300 ms. Logistic regression analysis revealed SSLs as an independent predictor for faster detection by endoscopists than by CADe (odds ratio: 6.1, 95% confidence interval: 4.6-807.3, P = 0.002).

CONCLUSION

Endoscopists ignored a substantial proportion of artifacts and successfully recognized most lesions detected by CADe. These findings provide insights into the CADe-endoscopist interaction. Nevertheless, SSL detection should be conducted with caution, even with CADe support.

TRIAL REGISTRATION

UMIN-CTR (UMIN000052891).

摘要

目的

计算机辅助检测系统（CADe）用于结肠镜筛查；然而，尚无研究定量分析内镜医师如何识别CADe识别出的候选部位。本研究的目的是确定内镜医师能够识别CADe检测部位的程度。

方法

这项前瞻性研究使用基于专用眼动追踪技术的系统，在结肠镜退出过程中记录单个视频内镜医师的眼睛位置和CADe检测部位。对所有视频进行回顾以计算内镜医师的识别率，其定义为CADe识别出的部位中，内镜医师凝视超过300毫秒的部位所占的百分比。使用逻辑回归模型确定内镜医师比CADe更快检测到病变的预测因素。

结果

在55例结肠镜检查中，CADe检测到447个部位（306个伪像、98个腺瘤、28个增生性息肉、12个无蒂锯齿状病变[SSLs]、3个黏膜内癌）。内镜医师的总体识别率为82.1%；腺瘤的识别率为95.9%，增生性息肉为92.8%，SSLs为100%，黏膜内癌为100%，伪像为75.8%。然而，追踪速度分析显示，内镜医师在300毫秒内仅追踪到33.3%由CADe检测到的SSLs。逻辑回归分析显示，SSLs是内镜医师比CADe更快检测到病变的独立预测因素（优势比：6.1，95%置信区间：4.6 - 807.3，P = 0.002）。