Department of Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong.
Department of Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong.
Gastrointest Endosc. 2020 Jul;92(1):11-22.e6. doi: 10.1016/j.gie.2020.02.033. Epub 2020 Feb 29.
We performed a meta-analysis of all published studies to determine the diagnostic accuracy of artificial intelligence (AI) on histology prediction and detection of colorectal polyps.
We searched Embase, PubMed, Medline, Web of Science, and Cochrane library databases to identify studies using AI for colorectal polyp histology prediction and detection. The quality of included studies was measured by the Quality Assessment of Diagnostic Accuracy Studies tool. We used a bivariate meta-analysis following a random-effects model to summarize the data and plotted hierarchical summary receiver operating characteristic curves. The area under the hierarchical summary receiver operating characteristic curve (AUC) served as an indicator of the diagnostic accuracy and during head-to-head comparisons.
A total of 7680 images of colorectal polyps from 18 studies were included in the analysis of histology prediction. The accuracy of the AI (AUC) was .96 (95% confidence interval [CI], .95-.98), with a corresponding pooled sensitivity of 92.3% (95% CI, 88.8%-94.9%) and specificity of 89.8% (95% CI, 85.3%-93.0%). The AUC of AI using narrow-band imaging (NBI) was significantly higher than the AUC using non-NBI (.98 vs .84, P < .01). The performance of AI was superior to nonexpert endoscopists (.97 vs .90, P < .01). For characterization of diminutive polyps using a deep learning model with nonmagnifying NBI, the pooled negative predictive value was 95.1% (95% CI, 87.7%-98.1%). For polyp detection, the pooled AUC was .90 (95% CI, .67-1.00) with a sensitivity of 95.0% (95% CI, 91.0%-97.0%) and a specificity of 88.0% (95% CI, 58.0%-99.0%).
AI was accurate in histology prediction and detection of colorectal polyps, including diminutive polyps. The performance of AI was better under NBI and was superior to nonexpert endoscopists. Despite the difference in AI models and study designs, AI performances are rather consistent, which could serve as a reference for future AI studies.
我们对所有已发表的研究进行了荟萃分析，以确定人工智能（AI）在预测和检测结直肠息肉组织学方面的诊断准确性。
我们检索了 Embase、PubMed、Medline、Web of Science 和 Cochrane 图书馆数据库，以确定使用 AI 进行结直肠息肉组织学预测和检测的研究。使用诊断准确性研究质量评估工具（Quality Assessment of Diagnostic Accuracy Studies tool）来衡量纳入研究的质量。我们使用双变量荟萃分析和随机效应模型来汇总数据，并绘制分层综合受试者工作特征曲线。分层综合受试者工作特征曲线下面积（area under the hierarchical summary receiver operating characteristic curve，AUC）作为诊断准确性的指标，并在头对头比较中使用。
共纳入 18 项研究中的 7680 张结直肠息肉图像进行组织学预测分析。AI 的准确性（AUC）为.96（95%置信区间 [CI]，.95-.98），对应的汇总敏感性为 92.3%（95% CI，88.8%-94.9%），特异性为 89.8%（95% CI，85.3%-93.0%）。窄带成像（narrow-band imaging，NBI）中 AI 的 AUC 显著高于非 NBI 的 AUC（.98 与.84，P<.01）。AI 的性能优于非专家内镜医师（.97 与.90，P<.01）。对于使用非放大 NBI 的深度学习模型对微小息肉进行特征描述，汇总阴性预测值为 95.1%（95% CI，87.7%-98.1%）。在检测息肉方面，汇总 AUC 为.90（95% CI，.67-1.00），敏感性为 95.0%（95% CI，91.0%-97.0%），特异性为 88.0%（95% CI，58.0%-99.0%）。
AI 在结直肠息肉的组织学预测和检测中具有较高的准确性，包括微小息肉。NBI 下 AI 的性能更好，且优于非专家内镜医师。尽管 AI 模型和研究设计存在差异，但 AI 的性能相当一致，可为未来的 AI 研究提供参考。
