基于深度学习的心脏影像学数据测量及其在心脏性猝死诊断中的应用  

Deep Learning-Based Cardiac Imaging Data Measurement and Its Application in Diagnosis of Sudden Cardiac Death

作　　者：李泽浩 刘宁国 董贺文 李廉杰 何恒辉 林丽华 刘茜[1] 杨明真 LI Ze-hao;LIU Ning-guo;DONG He-wen;LI Lian-jie;HE Heng-hui;LIN Li-hua;LIU Qian;YANG Ming-zhen(Department of Forensic Medicine,Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430030,China;Shanghai Key Laboratory of Forensic Medicine,Key Laboratory of Forensic Science,Ministry of Justice,Shanghai Forensic Service Platform,Academy of Forensic Science,Shanghai 200063,China)

机构地区：[1]华中科技大学同济医学院法医学系,湖北武汉430030 [2]司法鉴定科学研究院,上海市法医学重点实验室司法部司法鉴定重点实验室,上海市司法鉴定专业技术服务平台,上海200063

出　　处：《法医学杂志》2021年第4期546-554,共9页Journal of Forensic Medicine

基　　金：中央级公益性科研院所资助项目(GY2020Z-4,GY2021G-4);司法部司法鉴定重点实验室资助项目;上海市法医学重点实验室资助项目(21DZ2270800);上海市司法鉴定专业技术服务平台资助项目(19DZ2292700)。

摘　　要：在法医学领域,心脏性猝死的诊断受主观因素和人工测量方法的限制,使某些参数可能存在估计偏差或测量偏差。随着死后断层影像在死因鉴定和心脏病理学研究中发挥越来越重要的作用,应用深度学习等人工智能技术对海量的心脏影像学数据进行分析,为法医学鉴定和科研工作者对心脏疾病进行精细诊断和定量分析提供了可能。本文对近几年深度学习在心脏影像学领域的主要研究进行综述,为目前深度学习在心脏性猝死虚拟解剖中的应用提出可行的发展方向。In the field of forensic medicine,diagnosis of sudden cardiac death is limited by subjective factors and manual measurement methods,so some parameters may have estimation deviation or measurement deviation.As postmortem CT imaging plays a more and more important role in the appraisal of cause of death and cardiopathology research,the application of deep learning such as artificial intelligence technology to analyze vast amounts of cardiac imaging data has provided a possibility for forensic identification and scientific research workers to conduct precise diagnosis and quantitative analysis of cardiac diseases.This article summarizes the main researches on deep learning in the field of cardiac imaging in recent years,and proposes a feasible development direction for the application of deep learning in the virtual anatomy of sudden cardiac death at present.

关 键 词：法医病理学 医学影像技术学 心脏性猝死 虚拟解剖 深度学习 心血管系统 综述 

分 类 号：DF795.1[医药卫生—法医学]