基于集成学习的临床心电图分类算法研究

发布时间：2018-03-14 22:10

  本文选题：心电图　切入点：集成学习　出处：《生物医学工程学杂志》2016年05期 　论文类型：期刊论文

【摘要】：随着心电图数据量快速增长,计算机辅助心电图分析也有着越来越广阔的应用需求。本文在基于导联卷积神经网络的临床心电图分类算法上提出多种策略,进一步提升其在实际应用中的性能。首先用不同的预处理方法和训练方法获得两个不同的分类器,接着用多重输出预测法来增强每个分类器的性能,最后用贝叶斯方法进行融合。测试了超过15万条心电图记录,所提方法的准确率和受试者工作特征曲线下面积(AUC)分别为85.04%和0.918 5,明显优于基于特征提取的传统方法。
[Abstract]:With the rapid growth of ECG data, computer-aided ECG analysis has more and more extensive application needs. This paper proposes a variety of strategies on clinical ECG classification algorithm based on lead convolution neural network. Firstly, two different classifiers are obtained by using different preprocessing methods and training methods, and then the performance of each classifier is enhanced by using multiple output prediction method. Finally, more than 150,000 ECG records were tested by Bayesian method. The accuracy of the proposed method and the area under the operating characteristic curve were 85.04% and 0.918 5, respectively, which were obviously superior to the traditional method based on feature extraction.
【作者单位】： 中国科学院苏州纳米技术与纳米仿生研究所;中国科学院大学;
【分类号】：R540.41;TP391.4
，

本文编号：1613140