Premature Ventricular Contraction Detection from Ambulatory ECG Using Recurrent Neural Networks
Premature Ventricular Contraction Detection from Ambulatory ECG Using Recurrent Neural Networks
复制标题
DOI：
10.1109/embc.2018.8512858
复制
发表时间：
2018-07
期刊：
影响因子：
--
通讯作者：
Xue Zhou;Xin Zhu;Keijiro Nakamura;Noro Mahito
中科院分区：
文献类型：
--
作者：
Xue Zhou;Xin Zhu;Keijiro Nakamura;Noro Mahito
室性早搏（PVC）通常被认为是良性心律失常，在没有结构性心脏病。然而，频繁的早搏可能会显著增加心力衰竭的风险，甚至会因心肌病引起的心肌病而死亡。因此，高PVC计数被认为是预测重度心律失常风险的一种方法。可穿戴设备的发展为日常生活中过早收缩的检测提供了方便的工具。考虑到可穿戴设备记录的数据量巨大，需要开发可靠且低成本的数据分析程序，以实现真实的实时PVC检测。在这项研究中，我们使用具有长短期记忆的递归神经网络来检测PVC。通过与MIT-BIH心律失常数据库的验证，该方法的检测准确率为96%~ 99%。
Premature ventricular contraction (PVC) is usually considered to as benign arrhythmia in the absence of structural heart diseases. However, frequent premature beats may significantly increase the risk of heart failure and even death by an arrhythmia-induced cardiomyopathy. Therefore, high PVC counts have been considered as an approach to predict the risk of severe arrhythmias. Progress of wearable devices provides a convenient tool for the detection of premature contraction in casual life. Considering the huge quantities of data recorded by wearable devices, reliable and low-cost data analysis programs should be developed for real time PVC detection. In this research, we use recurrent neural networks with, long short-term memory to detect PVC. Through validating with MIT-BIH arrhythmia database, the detection accuracy of this method is 96%-99%.