超声内镜人工智能系统临床应用专家共识（2024，武汉）

Expert consensus on the clinical application of artificial intelligence system to endoscopic ultrasonography (2024, Wuhan)

摘要:

超声内镜是诊断肝胆胰系统病灶、黏膜下肿瘤等消化系统疾病、判断胃肠道早期癌浸润深度的有效工具。人工智能技术在超声内镜的质量控制和辅助诊断中起到了重要作用，但目前国内外尚无超声内镜人工智能系统临床应用的相关共识。2024年中华医学会消化内镜学分会大数据协作组组织全国领域内权威专家讨论，结合国内外最新循证医学证据，形成超声内镜人工智能系统临床应用专家共识，旨在为内镜医师应用超声内镜人工智能提供全面合理的决策证据。本共识包括人工智能在肝胆胰系统标准站点识别、消化道早期癌浸润深度预测、黏膜下肿瘤病理分型、肝胆胰异常病灶识别及病理分型等方面的9条推荐意见陈述。

Abstract:

Endoscopic ultrasonography is an effective tool for the diagnosis of digestive diseases including hepatobiliary and pancreatic lesions, submucosal tumors, and judgement of the depth of invasion of early gastrointestinal cancer. Artificial intelligence technology plays an important role in the quality control and auxiliary diagnosis of endoscopic ultrasonography, but there is no consensus on the application of artificial intelligence system to endoscopic ultrasonography at home and abroad. In 2024, Big Data Collaboration Group, Digestive Endoscopy Branch of Chinese Medical Association organized discussions among authoritative experts in the field across the country and formulated expert consensus on the clinical application of endoscopic ultrasonography artificial intelligence system based on the latest evidence-based medical evidence at home and abroad, aiming to provide endoscopists with comprehensive and reasonable decision-making evidence for the application of endoscopic ultrasonography artificial intelligence. This consensus included 9 recommendation statements regarding artificial intelligence in the identification of standard sites of the hepatobiliary and pancreatic system, prediction of infiltration depth of early digestive cancer, pathological classification of submucosal tumors and identification and pathological classification of abnormal lesions of the hepatobiliary and pancreatic systems.