Fuzzy Techniques with Aggregation Operators for Classification and Optimization of Epilepsy Risk Level from EEG Signals

Abstract

The purpose of this paper is to develop a fuzzy classification model for epilepsy risk level analysis from EEG signals. The parametric values such as energy, variance, and duration, covariance, positive and negative peaks, sharp and spike waves and events are derived in each epoch of two second duration in the EEG signal channels. Fuzzy techniques are used to classify the risk level in each epoch for all the channels, the risk level patterns obtained are found to have low values of sensitivity, specificity, performance index and quality value. In order to increase the classification rate, an optimization technique based on aggregation operators is used and a quality value of 23.78 is achieved when compared to the value 6.25 achieved in the previous case. A comparison of fuzzy techniques without and with optimization is studied. The focal epilepsy problem in normal fuzzy classification is solved using this new approach. A group of ten patients with known epilepsy findings are used in this study. Further research work can be carried out in the classification of epilepsy risk level of a long-term EEG signals about 3 minute's durations. The number of samples may be increased to improve the classification rate.

Indexing terms:

• EEG Signals
• Epilepsy
• Fuzzy logic
• Aggregation operators
• Risk levels

Additional information

Notes on contributors

R Hari Kumar

R Harikumar received his BE (ECE) degree from REC Trichy in 1988. He obtained his ME (Applied Electronics) degree from College of Engineering, Guindy, Anna University, Chennai in 1990. He has 15 years of teaching experience at college level. He worked as faculty in the Department of ECE at PSNA College of Engineering & Technology, Dindigul. He was Assistant Professor in IT at PSG College of Technology, Coimbatore. He also worked as Assistant Professor in ECE at Amrita Institute of Technology, Coimbatore. Currently, he is pursuing doctoral programme as a research scholar in Bio-medical Engineering under guidance of Dr. (Mrs) R Sukanesh Department of ECE, Thiagarajar College of Engineering, Madurai. He has published six papers in International and National Journals and also published around twenty two papers in International and National Conferences conducted both in India and abroad. His area of interest is Bio-signal Processing, Soft Computing, VLSI design and Communication Engineering. He is a life member of IETE and ISTE.

R Sukanesh

R Sukanesh senior faculty in Biomedical Engineering, TCE received her bachelor's degree in ECE from Government College of Technology Coimbatore in 1982. She obtained her ME (Communication Systems) degree from. PSG College of Technology, Coimbatore in 1985…and PhD in Biomedical Engineering from Madurai Kamaraj University, Madurai in 1999. Since 1985 she is working as a faculty in the Department of ECE at Thiagarajar College of Engineering, Madurai and presently, she is Assistant Professor of ECE in the same College. Her main research area includes Biomedical Instrumentation, Neural Networks, Bio Signal Processing and Mobile Communication. She is guiding twelve PhD theses in these areas. She has published several papers in International and National Journals and also published around sixty papers in International and National Conferences conducted both in India and abroad. She has delivered a number of invited lectures in various universities. She has a diploma in Higher Learning and has co-authored a book on Gandhian Thoughts. She is life member of Bio=medical Society of India, Indian Association of Bio-medical Scientist and ISTE.d the Chairman of the ISTE-IIT Bombay Chapter (2001–06).