摘要
在心脑血管领域,医学影像与机器学习(machine learning,ML)的结合能够在影像数据中实现信息的更深入挖掘,并辅助疾病的分析诊断。目前,ML已经在CT成像技术、磁共振成像技术、超声成像技术、核医学成像技术等医学影像技术上有所应用,减少了不同个体主观判断所产生的误差,同时提高了诊疗效率。我们总结了心脑血管诊疗领域常用医学影像技术的特点,分别从心脑血管影像的分类、检测、分割和生成四方面对ML的应用进行探讨,并对ML在心脑血管影像分析领域的未来应用方向进行展望。
In the field of cardiovascular and cerebrovascular,medical imaging combines with machine learning(ML)can realize deeper information mining in image data and assist disease analysis and diagnosis.At present,ML has been applied to medical imaging technologies such as CT imaging technology,magnetic resonance imaging technology,ultrasound imaging technology and nuclear medicine imaging technology,etc.It reduces the errors caused by the subjective judgment of different individuals and improves the efficiency of diagnosis and treatment.We summarize the characteristics of medical imaging technologies commonly used in the field of cardiovascular and cerebrovascular diagnosis and treatment,and discuss the application of machine learning in the four aspects of classification,detection,segmentation and generation of cardiovascular and cerebrovascular images,then prospect the future application of machine learning in the field of cardiovascular and cerebrovascular image analysis.
作者
高天欣
褚天琪
张栩阳
梅玉倩
陈端端
GAO Tianxin;CHU Tianqi;ZHANG Xuyang;MEI Yuqian;CHEN Duanduan(School of Life Science, Beijing Institute of Technology, Beijing 100081, China)
出处
《生物医学工程研究》
2021年第2期197-202,共6页
Journal Of Biomedical Engineering Research
基金
国家自然科学基金资助项目(81970404、81911530224、81601561)
国家重点研发计划项目(2018AAA0102600)
北京市自然科学基金资助项目(Z190014、L192010)
北京市科技新星支持计划项目(Z181100006218008)。
关键词
机器学习
心脑血管疾病
医学影像
图像分析
神经网络
Machine learning
Cardiovascular and cerebrovascular diseases
Medical imaging
Image analysis
Neural network
