北京生物医学工程

基于抗混叠小波变换的胎儿心电信号分离方法

Fetal ECG separation method based on anti-aliasing wavelet transform

作者：王旭蔡宗平林生佐

单位：广东环境保护工程职业学院 (广东佛山 528216) 通信作者:王旭 E-mail:954582137@ qq. com

关键词：抗混叠; 小波变换; 胎儿心电信号; 母体心电信号; 心电峰值

分类号：R318;TP301. 6

出版年·卷·期（页码）：2020·39·4（398-405）

摘要：

目的胎儿心电图能够较好地反映胎儿在子宫内的发育状况,但是由于采集的胎儿心电信号中混有噪声干扰,给医学诊断带来极大干扰。抗混叠小波变换算法能够从混有噪声干扰的源信号中提取胎儿心电信号,且当胎儿心电信号与母体心电信号混叠时,该方法仍能够提取胎儿心电信号。基于此,本文提出一种基于抗混叠小波变换的胎儿心电信号分离方法。方法首先对原始心电信号进行滤波预处理,再利用小波变换分离母体心电信号和胎儿心电信号,最后根据抗混叠分离算法获取混合心电信号中的胎儿心电信号,得到满周期的胎儿心电信号。结果该方法能够较好地获取胎儿心电波形,胎儿心电波形识别准确率可达 100%,在信噪比较低的情况下,识别准确率仍可达到 77. 78%。应用此算法在国外 MIT-BIT 心电信号数据和国内医院临床心电信号数据中进行实验仿真,并与先前学者的胎儿心电信号提取方法进行对比。结论此方法具有较高的识别准确率以及在临床应用中的可靠性和可行性。

Objective Fetal electrocardiogram can reflect the development of fetus in the womb better, but the interference of noise in the collected fetal ECG signals brings great interference to the medical diagnosis. The anti-aliasing wavelet transform algorithm can extract the fetal ECG signals from the source signals with noise interference,and when the fetal ECG signals are aligned with the maternal ECG signals,the method can still extract the fetal ECG signals. And on account of which,this paper proposes a method of fetal ECG separation based on anti-aliasing wavelet transform. Methods First,the ECG signals are processed by noise filtering. Then,we use wavelet transform to separate maternal ECG and fetal ECG. Finally,the fetal ECG signals in the mixed ECG are obtained by using the anti-aliasing separation algorithm. Results This method can obtain the ECG waveform better,and the accuracy of fetal ECG signal recognition can be 100%. In the case of low signal-to-noise ratio,the accuracy of fetal ECG signal recognition is still 77. 78%. The algorithm is simulated in foreign MIT-BIT ECG database and clinical ECG data in domestic hospitals and is compared with the previous methods of fetal ECG extraction. Conclusions The high recognition accuracy of this method,as well as its reliability and feasibility in clinical application is proved.

参考文献：

[ 1 ] Dessì A,Pani D,Raffo L. An advanced algorithm for fetal heart rate estimation from non-invasive low electrode density recordings [J]. Physiological Measurement,2014,35 (8):1621.

[ 2 ] John RG,Ramachandran KI,Nair RH,et al. Portable fental ECG extractor from abdECG [ J ]. International Conference on Communication and Signal Processing,2016:845-848.

[ 3 ] Singh R,Singh A,Kaur J. Adaptive filter design for extraction of fetus ECG signal [M]. Springer India:Proceedings of the Second International Conference on Computer and Communication Technologies,2015:109-117.

[ 4 ] Turnip M,Saragih RIE,Dharma A,et al.Extraction of ECG signal with adaptive filter for hearth abnormalities detection[J]. Journal of Physics:Conference Series,2018,1007:012019.

[ 5 ] Lin CS, Hasting E. Blind source separation of heart and lung sounds based on nonnegative matrix factorization [ C ] / / International Symposium on Intelligent Signal Processing and Communications Systems. Naha-shi, Japan: IEEE Press, 2013: 731-736.

[ 6 ] Li R,Chen BF. FECG extraction algorithm based on BSS using temporal structure and DWT [ J ]. Applied Mechanics and Materials,2014,571-572:209-212.

[ 7 ] Sait NA,Thangarajan M,Snehalatha U. Neural network based on Verilog HDL for fetal ECG extraction [ J]. International Journal of Biomedical Research,2016,7(10):698-701.

[ 8 ] 袁延超,吴水才,袁丽,等. 基于 BP 神经网络的胎儿心电提取算法研究[J]. 生物医学工程与临床,2018,22(3):237-243.

Yuan YC, Wu SC, Yuan L, et al. Research of fetal electrocardiogram extraction algorithm based on BP neural network [ J ]. Biomedical Engineering and Clinical Medicine,2018,22(3):237-243.

[ 9 ] Fu Y, Xiang S, Chen T, et al. Noninvasive extraction of fetal electrocardiogram based on Support Vector Machine [ C ] / /Applied Optics and Photonics China. Beijing,China:International Society for Optics and Photonics,2015:96720O.

[10] Sarmiento LO,Gonzalez A,Roig JM. Synthetic database for testing algorithms of fetal ECG extraction from abdominal ECG [ C] / /XVII Symposium of Image, Signal Processing, and Artificial Vision. Antioquia,Colombia:IEEE Press,2012:56-61.

[11] 华晶. 可穿戴式系统心电信号压缩感知与心律失常分类研究[D]. 南昌:南昌大学,2017:24-35.

[12] 刁晨. 心电信号质量评估与心率变异性分析方法研究[ D].兰州:兰州理工大学,2016:23-26.

[13] 曹细武,邓亲恺. 心电图各波的频率分析[ J]. 中国医学物理学杂志,2001,18(1):46-48.

[14] Sameni R,Clifford GD. A review of fetal ECG signal processing issues and promising directions [ J ]. The Open Pacing,Electrophysiology & Therapy Journal,2010,3:4-20.

[15] Castillo E,Morales DP,Botella G,et al. Efficient wavelet-based ECG processing for single-lead FHR extraction [ J ]. Digital Signal Processing,2013,23 (6):1897-1909.

[16] Donoho DL,Johnstone IM. Adapting to unknown smoothness via wavelet shrink-age [ J ]. Journal of the American Statistical Association,1995,90(432):1200-1224.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录