基于深度因子分析的动态心电信号降噪算法研究

发布时间：2018-05-27 13:32

本文选题：动态心电图 + 远程心电监护　；参考：《河北大学》2017年硕士论文

【摘要】：近年来,我国人口老龄化和城镇化进程正在加速,越来越多的人受到心脏疾病的困扰。针对患者有明显的自觉症状,但是静态心电图难以捕捉到有效的诊断依据,医生会建议进行动态心电监测。在远程医疗背景下,去除动态心电信号中的噪声,提高远程心电监护系统自动检测的准确率,逐渐成为研究的重点和难点。曾被广泛应用于语音识别和人脸识别中的因子分析为心电信号处理提供了新的切入点,为了去除心电信号中的复杂噪声,本文提出了一种基于深度因子分析的降噪算法。论文主要内容如下:(1)构建因子分析的心电信号降噪模型。将因子分析模型引入到心电信号降噪领域中,基于因子分析模型与心电信号数据库,通过机器学习获取心电信号的隐因子,并将噪声区分出来。(2)提出一种基于深度因子分析的心电信号降噪算法。考虑到心电信号的隐因子受到噪声的干扰,通过对隐因子逐层加深构建因子分析模型的方法,将每层的隐因子作为输入,获取深层的隐因子,丢弃每层的高斯噪声,最后由顶层隐因子自顶向下重构出降噪后的心电信号,实现心电信号的降噪。结果表明,基于深度因子分析的心电信号降噪算法取得了较好的去噪效果。(3)将深度因子分析降噪算法应用到心电监控平台。为了验证本文的研究成果,将深度因子分析降噪算法应用于团队自主研发的智慧心电监测平台中。由在平台上抽取的用户数据结果表明,本文提出的心电信号降噪算法可以在滤除复杂噪声的同时保持心电信号的主要特征波形。
[Abstract]:In recent years, population aging and urbanization are accelerating, more and more people suffer from heart disease. For patients with obvious symptoms, but static electrocardiogram is difficult to capture effective diagnostic basis, doctors will recommend dynamic ECG monitoring. In the context of telemedicine, removing noise from dynamic ECG signals and improving the accuracy of automatic detection of remote ECG monitoring system have gradually become the focus and difficulty of the research. Factor analysis, which has been widely used in speech recognition and face recognition, provides a new entry point for ECG signal processing. In order to remove the complex noise in ECG signal, a denoising algorithm based on depth factor analysis is proposed in this paper. The main contents of this paper are as follows: (1) construct the ECG denoising model based on factor analysis. Factor analysis model is introduced into the field of ECG signal denoising. Based on factor analysis model and ECG database, the hidden factors of ECG signal are obtained by machine learning. This paper proposes a denoising algorithm for ECG signal based on depth factor analysis. Considering that the hidden factor of ECG signal is disturbed by noise, the hidden factor of each layer is taken as input to get the deep hidden factor, and the Gao Si noise of each layer is discarded by the method of constructing the factor analysis model layer by layer. Finally, the top hidden factor is used to reconstruct the denoised ECG signal from top to bottom to realize the de-noising of ECG signal. The results show that the de-noising algorithm of ECG signal based on depth factor analysis has achieved better denoising effect. The depth factor analysis de-noising algorithm is applied to ECG monitoring platform. In order to verify the research results of this paper, the depth factor analysis (DFA) denoising algorithm is applied to the intelligent ECG monitoring platform developed independently by the team. The results of the user data extracted on the platform show that the proposed ECG denoising algorithm can filter the complex noise while maintaining the main characteristic waveform of the ECG signal.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R540.4;TN911.4

【相似文献】