摘要
综述了深度学习在单导联和多导联心电图(electrocardiograph,ECG)心律失常检测中的应用现状,分析了深度学习在心律失常检测应用中存在泛化能力差、可解释性差、时间复杂度大等问题,并提出了解决方案。指出了随着算法的不断迭代与更新、数据集的增加以及硬件设备性能的提升,深度学习在ECG心律失常检测中的应用前景更加广阔。
The current situation of deep learning applied to single-and multi-lead ECG detection of arrhythmia was reviewed.The problems of deep learning during the application in generalization,interpretability and time complexity were analyzed,and the countermeasures were put forward accordingly.It's pointed out deep learning would be applied widely in arrhythmia ECG detection with the development of the algorithm,dataset and hardware.
作者
黄丽
蔡定建
凌世康
欧阳昊
李嘉
HUANG Li;CAI Ding-jian;LING Shi-kang;OUYANG Hao;LI Jia(Faculty of Intelligent Manufacturing,Wuyi University,Jiangmen 529020,Guangdong Province,China)
出处
《医疗卫生装备》
CAS
2024年第2期105-112,共8页
Chinese Medical Equipment Journal
基金
江门市科技计划项目(2023JC01005)。
