Efficient Removal of Artifacts from EEG SIGNAL Using Enhanced Hybrid Learning Method

Q2 Social Sciences

Studies on Ethno-Medicine Pub Date : 2017-10-02 DOI:10.1080/09735070.2017.1385937

B. Paulchamy

{"title":"Efficient Removal of Artifacts from EEG SIGNAL Using Enhanced Hybrid Learning Method","authors":"B. Paulchamy","doi":"10.1080/09735070.2017.1385937","DOIUrl":null,"url":null,"abstract":"ABSTRACT In this paper, the analysis and removal of artifacts is done by the proposed technique. Normally, ECG is one of the components of artifacts source and EEG is mixed by various artifacts and affects the electroencephalographic data. For further clinical analysis the data preparation is important to minimize the artifacts. In proposed method, Improved Adaptive Neuro-Fuzzy Inference System (IANFIS) and Improved ANFISParticle Swarm Optimization (IANFIS-PSO) algorithms are used to separate the signals of ECG and EEG for eliminating artifacts and to intensify the estimation of EEG signal quality. The pre-processing is done by ennobled quantum based genetic algorithm for fast process of optimization and removal of noise interference. The simulation result shows the improvement in Signal-to-Noise Ratio (SNR), minimum Mean-Square Error (MSE) along with the Power Spectrum Density (PSD) plot, which are used to measure the performance comparison of proposed with existing algorithm. The prospective method performs with more appropriate process of enhanced hybrid learning method and outperforms in minimizing the artifacts of ECG from the corrupted signals of EEG.","PeriodicalId":39279,"journal":{"name":"Studies on Ethno-Medicine","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2017-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/09735070.2017.1385937","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Studies on Ethno-Medicine","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/09735070.2017.1385937","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Social Sciences","Score":null,"Total":0}

引用次数: 2

Abstract

ABSTRACT In this paper, the analysis and removal of artifacts is done by the proposed technique. Normally, ECG is one of the components of artifacts source and EEG is mixed by various artifacts and affects the electroencephalographic data. For further clinical analysis the data preparation is important to minimize the artifacts. In proposed method, Improved Adaptive Neuro-Fuzzy Inference System (IANFIS) and Improved ANFISParticle Swarm Optimization (IANFIS-PSO) algorithms are used to separate the signals of ECG and EEG for eliminating artifacts and to intensify the estimation of EEG signal quality. The pre-processing is done by ennobled quantum based genetic algorithm for fast process of optimization and removal of noise interference. The simulation result shows the improvement in Signal-to-Noise Ratio (SNR), minimum Mean-Square Error (MSE) along with the Power Spectrum Density (PSD) plot, which are used to measure the performance comparison of proposed with existing algorithm. The prospective method performs with more appropriate process of enhanced hybrid learning method and outperforms in minimizing the artifacts of ECG from the corrupted signals of EEG.

查看原文本刊更多论文

基于增强混合学习方法的脑电信号伪影有效去除

在本文中，利用提出的技术进行伪影的分析和去除。通常情况下，心电是伪影源的组成部分之一，而脑电图是由各种伪影混合而成，影响着脑电图数据。对于进一步的临床分析，数据准备对于最小化伪影非常重要。该方法采用改进的自适应神经模糊推理系统(IANFIS)和改进的anfiss粒子群算法(IANFIS- pso)分离心电和脑电图信号，消除伪像，加强对脑电信号质量的估计。预处理采用基于赋能量子的遗传算法进行，以实现快速优化过程和去除噪声干扰。仿真结果表明，该算法在信噪比(SNR)、最小均方误差(MSE)和功率谱密度(PSD)图上均有改善，并可用于衡量该算法与现有算法的性能比较。前瞻性方法采用了更合适的增强混合学习方法，在最小化脑电信号干扰中产生的心电伪影方面表现优异。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Studies on Ethno-Medicine Social Sciences-Cultural Studies

CiteScore

0.50

自引率

0.00%

发文量

期刊介绍： Studies on Ethno-Medicine is a peer reviewed, internationally circulated journal. It publishes reports of original research, theoretical articles, timely reviews, brief communications, book reviews and other publications in the interdisciplinary field of ethno-medicine. The journal serves as a forum for physical, social and life scientists as well as for health professionals. The transdisciplinary areas covered by this journal include, but are not limited to, Physical Sciences, Anthropology, Sociology, Geography, Life Sciences, Environmental Sciences, Botany, Agriculture, Home Science, Zoology, Genetics, Biology, Medical Sciences, Public Health, Demography and Epidemiology. The journal publishes basic, applied and methodologically oriented research from all such areas. The journal is committed to prompt review, and priority publication is given to manuscripts with novel or timely findings, and to manuscript of unusual interest. Further, the manuscripts are categorised under three types, namely - Regular articles, Short Communications and Reviews. The researchers are invited to submit original papers in English (papers published elsewhere or under consideration elsewhere shall not be considered).