A Multi-View SVM Approach for Seizure Detection from Single Channel EEG Signals

Abstract

Seizures are the part of the epilepsy that occurs in central nervous system which leads to abnormal brain activity. Electroencephalogram (EEG) signal recordings are mostly used in epileptic seizure detection process. Detection of seizures is a crucial part for further treatment of patients. This paper proposes a multi-view SVM model for seizure detection using the single channel EEG signals. In this experiment, two views of the EEG data have been extracted, (1) the time domain features using Independent Component Analysis (ICA) and (2) power spectral densities are obtained in the frequency domain. Extracted features have been fed to multi-view SVM classification model. In this study, a single channel EEG dataset is used for seizure detection. Performance estimation parameters namely Accuracy, Sensitivity, Specificity, F1-score, and AUC value have been estimated for evaluating the proposed model. The model classified seizure and non-seizure over the sets A vs E and B vs E with an accuracy greater than 99% using k-fold cross validation. The classification accuracy obtained by multi-view SVM is better by 1–4% than single view SVM using the same features. Furthermore, the proposed model is also compared with existing single view SVM models. It is observed that the multi view SVM model performed significantly better, compare to a single view SVM model over the same features.

Keywords:

• Electroencephalogram
• EEG signals
• Epilepsy
• Independent Component Analysis (ICA)
• Power Spectral Density (PSD)
• Seizure
• Support vector machine (SVM)

ACKNOWLEDGEMENT

This work was carried out at Interactive Technologies & Multimedia Research Lab (ITMR Lab) supported by the Department of Information Technology, Indian Institute of Information Technology Allahabad (https://www.iiita.ac.in/), UP, India. The authors are grateful for this support.

Additional information

Notes on contributors

Gopal Chandra Jana

Gopal Chandra Jana currently pursuing PhD degree in the Department of Information Technology, IIIT Allahabad. He is associated with Interactive Technologies & Multimedia Research (ITMR) Lab, IIIT Allahabad. He received his BE degree in information technology from University Institute of Information Technology, Burdwan University, West Bengal, India and MTech degree in computer science and engineering from KIIT deemed to University, Bhubaneswar, Odisha, India. His research interest includes EEG signal analysis, machine learning applications in non-invasive neuro-imaging modalities. Corresponding author. E-mail: [email protected], [email protected]

Mogullapally Sai Praneeth

Mogullapally Sai Praneeth was summer research intern at Interactive Technologies & Multimedia Research Lab (ITMR Lab), IIIT Allahabad. He was born in Hyderabad, India, in 2000. He is currently pursuing BTech degree in electronics and communication engineering from Indian Institute of Technology, Bhubaneswar. His research interests are in the area of artificial intelligence, self-supervised learning, computer vision, and machine learning. E-mail: [email protected]

Anupam Agrawal

Anupam Agrawal is working as a professor of computer science and information technology at Indian Institute of Information Technology Allahabad (IIIT-A) since the year 2012. Before joining IIIT-A in the year 2000, he was working as scientist “D” at DEAL, DRDO, Govt. of India, Dehradun where he had contributed significantly in various national level projects. He is head of the “Interactive Technologies and Multimedia Research” (ITMR) Lab. which he had setup at IIIT-A with funding from a MHRD-sponsored R&D project (2003) as principal investigator. He received his MSc degree in computer science from JK Institute of Applied Physics & Technology, University of Allahabad, Allahabad, MTech degree in computer science and engineering from Indian Institute of Technology (IIT) Madras, Chennai and PhD degree in Information Technology from Indian Institute of Information Technology Allahabad (in association with Indian Institute of Technology, Roorkee). He was a postdoctoral researcher at the Department of Computer Science & Technology, University of Bedfordshire (UK) during which he had contributed significantly in two major European Commission projects which were rated ‘excellent’ by the commission. He is a Senior Member of IEEE (USA), Member of ACM(USA), and a Fellow of IETE (India). His research interests include computer vision, signal and image processing, visual computing, machine learning, remote sensing, soft computing and human-computer interaction. He is the recipient of Er. Hari Mohan Memorial Award for the best research paper by the Institution of Engineers (India), Lucknow in November, 2005 and other similar awards subsequently. E-mail: [email protected]