Artificial intelligence-based diagnosis of abnormalities in small-bowel capsule endoscopy
Artificial intelligence-based diagnosis of abnormalities in small-bowel capsule endoscopy
复制标题
DOI：
10.1055/a-1881-4209
复制
发表时间：
2022-08-05
期刊：
影响因子：
9.3
通讯作者：
Lin, Rong
中科院分区：
文献类型：
--
作者：
Ding, Zhen;Shi, Huiying;Lin, Rong
背景有必要进一步开发基于深度学习的人工智能（AI）技术来自动诊断小肠胶囊内窥镜（SBCE）视频中的多种异常。我们的目的是开发一个AI模型，与不同经验水平的医生比较其诊断性能，并进一步评估其对医生诊断SBCE视频中的多种异常的辅助作用。方法使用2565名患者的280426张图像对AI模型进行训练，并在240个视频中验证诊断性能。结果AI模型对红斑、炎症、血含量、血管病变、突出性病变、寄生虫、憩室和正常变异分别为 97.8%、96.1%、96.1%、94.7%、95.6%、100%、100% 和 96.4%。特异性分别为86.0%、75.3%、87.3%、77.8%、67.7%、97.5%、91.2%和81.3%。准确率分别为95.0%、88.8%、89.2%、79.2%、80.8%、97.5%、91.3%和93.3%。对于初级医生来说，人工智能模型的辅助将整体准确率从 85.5% 提高到 97.9%（P < 0.0125，Bonferroni 校正）。 结论 这种训练有素的人工智能诊断模型基于视频级识别同时自动诊断多个小肠异常，对于经验不足的内窥镜医生来说，有潜力成为一个优秀的辅助系统。
Background Further development of deep learning-based artificial intelligence (AI) technology to automatically diagnose multiple abnormalities in small-bowel capsule endoscopy (SBCE) videos is necessary. We aimed to develop an AI model, to compare its diagnostic performance with doctors of different experience levels, and to further evaluate its auxiliary role for doctors in diagnosing multiple abnormalities in SBCE videos.Methods The AI model was trained using 280426 images from 2565 patients, and the diagnostic performance was validated in 240 videos.Results The sensitivity of the AI model for red spots, inflammation, blood content, vascular lesions, protruding lesions, parasites, diverticulum, and normal variants was 97.8%, 96.1 %, 96.1%, 94.7%, 95.6%, 100%, 100%, and 96.4%, respectively. The specificity was 86.0%, 75.3%, 87.3%, 77.8%, 67.7%, 97.5%, 91.2%, and 81.3%, respectively. The accuracy was 95.0%, 88.8%, 89.2%, 79.2%, 80.8%, 97.5%, 91.3%, and 93.3%, respectively. For junior doctors, the assistance of the AI model increased the over- all accuracy from 85.5% to 97.9% (P 0.0125, Bonferroni corrected).Conclusions This well-trained AI diagnostic model automatically diagnosed multiple small-bowel abnormalities simultaneously based on video-level recognition, with potential as an excellent auxiliary system for less-experienced endoscopists.