影像组学和深度学习在肝硬化食管-胃底静脉曲张中的研究进展  

Research progress of radiomics and deep learning in esophagogastric varices in patients with liver cirrhosis

作　　者：赵顺敏 许永生[1,2,3] 黎金葵 杨正高 雷军强 Zhao Shunmin;Xu Yongsheng;Li Jinkui;Yang Zhenggao;Lei Junqiang(The First Clinical Medical School,Lanzhou University,Lanzhou 730000,China;Department of Radiology,the First Hospital of Lanzhou University,Lanzhou 730000,China;Intelligent Imaging Medical Engineering Research Center of Gansu Province,Lanzhou 730000,China)

机构地区：[1]兰州大学第一临床医学院,甘肃兰州730000 [2]兰州大学第一医院放射科,甘肃兰州730000 [3]甘肃省智能影像医学工程研究中心,甘肃兰州730000

出　　处：《兰州大学学报（医学版）》2023年第3期71-75,82,共6页Journal of Lanzhou University（Medical Sciences）

基　　金：兰州市城关区科技计划资助项目(2020RCCX0053)。

摘　　要：食管-胃底静脉曲张是肝硬化门静脉高压的常见并发症,食管-胃底静脉曲张破裂出血是临床常见的急重症,死亡率极高,对食管-胃底静脉曲张检测和评估可以更好地进行针对性治疗,降低消化道出血的发生率和死亡率。影像组学和深度学习作为支持临床决策和精准医学的方法,能够从医学图像中提取高通量和定量特征并加以数据挖掘和统计分析,揭示这些特征与疾病之间的相关性,可用于准确预测肝硬化患者门静脉高压和食管-胃底静脉曲张。本综述主要介绍了影像组学和深度学习技术在食管-胃底静脉曲张中的无创诊断、分级和预后方面的最新进展。Esophagogastric varix is a common complication of cirrhotic portal hypertension.Esophagogastric varix bleeding is a common clinical emergency with a high mortality.The detection and evaluation of esophagogastric varices can facilitate better targeted treatment to reduce the incidence and mortality of gastrointestinal bleeding.As methods to support clinical decision-making and accurate medicine,radiomics and deep learning can extract high-throughput and quantitative features from medical images and reveal the correlation between these features and diseases by data mining and statistical analysis.It can be used to accurately predict portal hypertension detection and esophagogastric varices in patients with liver cirrhosis.This review mainly introduced the latest progress of radiomics and deep learning techniques in non-invasive diagnosis,grading and prognosis of esophagogastric varices.

关 键 词：深度学习 影像组学 食管-胃底静脉曲张 肝硬化 门静脉高压 

分 类 号：R575.2[医药卫生—消化系统]