Analysing the Inclusion of Soft Computing Techniques in Denoising EEG Signal
Ashish Raj1, Akanksha Deo2, Mangesh S. Tomar3, Manoj Kumar Bandil4
1Ashish Raj, Assistant.Prof, Department of Electrical Engineering, IITM, Gwalior India.
2Akanksha Deo, Assistant.Prof, Department Of Electrical Engineering, IITM, Gwalior India.
3Mangesh Tomar, Assistant.Prof, Department Of Electrical Engineering, IITM, Gwalior India.
4Manoj Kumar Bandil, Associate Prof, Department Of Electrical Engineering, IITM, Gwalior India.
Manuscript received on September 01, 2012. | Revised Manuscript received on September 02, 2012. | Manuscript published on September 05, 2012. | PP: 107-112 | Volume-2 Issue-4, September 2012. | Retrieval Number: D0909072412/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The electrical nature of the human nervous system has been innovated for more than a century. It is prominent that the variation of the surface potential distribution on the scalp reflects function and activities emerging from the underlying brain. This variation of the surface potential can be recorded by placing an array of electrodes to the scalp, and measuring the voltage between pairs of these electrodes. These measured voltages are then filtered, amplified, and recorded. The resulting data is called the EEG. As per the usefulness, EEG has proved to be an important tool for diagnosis, monitoring and managing various nervous disorders. The electrical activity of brain changes in accordance with various parameters inside & outside environment. A number of severe disorders in human body which were impossible to be traced in early stages are easily being signal processing stages are being predicted with help of EEG. But there are certain artifacts which are present in raw EEG recording. These raw signals are firstly processed with help of mathematical tools in order to make them more and more informative. The informative signal thus calculated from recording is known as ERP (event related potential). These ERP data are very specific and changes with every physiological & biological change in human body. Thus the analysis of ERP has got numerous clinical importance.But there are certain artifacts which are present in raw EEG recording. These artifacts make the ERP contaminated and it introduces inconsistency in the output. These artifacts in EEG signals arise due to two types of factors; Biological factors and External factors. The Biological factors are caused by EOG (Elecro-oculogram), ECG (Electrocardiogram), EMG (Electromyogram) and Respiratory (PNG).The external factors are caused due to line-interference, leads and electrodes. These noises have an adverse effect on EEG signals and act as a contamination to obtain clear cut information from EEG signals .Thus it is perquisite to eliminate these artifacts from the EEG. The ERP generated from artifacts free EEG are most suitable for versatile researches and efficient diagnosis. The clinical information thus obtained is of considerable importance in identifying different pathologies. Thus artifact rejection is most important preliminary stage before ERP analysis. This is a paper scrutinizing different soft computing methods for removing artifacts with illustrating characteristics of a good informative EEG signal. In this paper we have discussed about inclusion of several soft computing techniques with the conventional artifact removal approaches.
Keywords: EEG; EMG; ECG; ocular artifacts; muscular artifacts; spike detection; Wavelet transform; Neural network., Fuzzy logic; Genetic Algorith