Brain Seizure Detection and Classification Using EEG Signals

By Varsha K. Harpale and Vinayak Bairagi

Brain Seizure Detection and Classification Using Electroencephalographic Signals presents EEG signal processing and analysis with high performance feature extraction. The book covers the feature selection method based on One-way ANOVA, along with high performance machine learning classifiers for the classification of EEG signals in normal and epileptic EEG signals. In addition, the authors also present new methods of feature extraction, including Singular Spectrum-Empirical Wavelet Transform (SSEWT) for improved classification of seizures in significant seizure-types, specifically epileptic and Non-Epileptic Seizures (NES). The performance of the system is compared with existing methods of feature extraction using Wavelet Transform (WT) and Empirical Wavelet Transform (EWT).

The book's objective is to analyze the EEG signals to observe abnormalities of brain activities called epileptic seizure. Seizure is a neurological disorder in which too many neurons are excited at the same time and are triggered by brain injury or by chemical imbalance.

• Presents EEG signal processing and analysis concepts with high performance feature extraction
• Discusses recent trends in seizure detection, prediction and classification methodologies
• Helps classify epileptic and non-epileptic seizures where misdiagnosis may lead to the unnecessary use of antiepileptic medication
• Provides new guidance and technical discussions on feature-extraction methods and feature selection methods based on One-way ANOVA, along with high performance machine learning classifiers for classification of EEG signals in normal and epileptic EEG signals, and new methods of feature extraction developed by the authors, including Singular Spectrum-Empirical Wavelet

• Language: English
• Publisher: Academic Press
• Release date: Sep 9, 2021
• ISBN: 9780323911214

Varsha K. Harpale

Dr. Mrs. Varsha K. Harpale, is working as academician in Pimpri Chinchwad College of Engg. Pune Maharashtra, India. Her work profile is Associate Professor in E&TC and Associate Dean Quality Assurance for maintaining and improving quality of education in autonomous institute. She has completed her PhD in biomedical signal processing. She has 20 years of teaching experiences, 2 books , 4 copyrights, 2 patents, best paper awards and 30+ quality publication on her credits. She is currently working as member secretary of IEEE Signal Processing Society, Pune Chapter.

Book preview

Chapter 1: Introduction

Engineering research in neurological disorders

Abstract

A study conducted by the World Health Organization concluded that the global health impact of neurological disorders had been underestimated. Providing a rough estimate, 30 million people suffer from neurological disorders. The incidence of neurological disorders has been projected to increase by 12% between 2005 and 2030 [1]. The prevalence rates of neurological disorders range from 967 to 4070 (mean, 2394) per 100,000 inhabitants from different regions of India, providing a rough estimate of over 30 million people with neurological disorders. The objective of this chapter is to review recent engineering research trends in neurological disorders and epilepsy. We will focus on the engineering research in seizure detection, prediction, and classification to highlight opportunities in the field.

Keywords

epilepsy; nonepileptic seizures (NES); psychogenic nonepileptic seizures (PNES)

1.1 Introduction

Prevalence rates of common neurological disorders including epilepsy, stroke, Parkinson's Disease (PD), and tumors were estimated through a survey conducted across different regions of the world. Considering all variations of 15 neurologic conditions, i.e., Alzheimer disease and related dementia, amyotrophic lateral sclerosis, brain tumors, cerebral palsy, dystonia, epilepsy, Huntington disease, hydrocephalus, multiple sclerosis, muscular dystrophy, PD, spinal cord injury, Tourette syndrome, and traumatic brain injury the surveys are conducted by standard medical agency such as WHO. Fig. 1.1 shows comparative technical research conducted every year for the identification, prevention, and risk analysis of most of the neurological disorders. It summarizes an on-line study published by IEEE Explore and Science Direct for 2017–2018. With respect to the massive burden associated with neurological disorders, it is clear that the neurological services and resources were disproportionately scarce, especially in low-income and developing countries.

Figure 1.1 Technological research conducted per year in neurological disorders.

The location-based prevalence of neurological disorders is evaluated mainly based on disability-adjusted life in years (DALYs), years of life lost (YLLs), and years lived with disability (YLDs), as stated by the Global Burden of Disease (GBD) study [2]. Statistically, it is observed that 10.2% of global DALYs and 16.8% of global deaths are due to neurological disorders, and between 1990 and 2015, the number of fatalities increased by 36.7%, and DALYs increased by 7.4%. Fig. 1.2 shows the severity of neurological diseases as compared to other disorders affecting human life, and it is observed that 6.3% of total sufferers have a neurological disorder. Thus, 6.3 million people die every year due to neurological disorders [2]. Neurological disorders have a significant impact on disability and mortality worldwide [3]. Epilepsy is also a high-prevalence disease, accounting for 5% of DALYs and 1.3% of deaths. In 2016, 45.9 million patients had epilepsy; the most affected age groups were 5–9 years and >80 years of age. The DALY rate in men was found higher than in women [4]. The World Health Organization, the International League Against Epilepsy, and the International Bureau for Epilepsy operate worldwide, providing technical information about epilepsy and its consequences. They also assist governments and those suffering from epilepsy to reduce the burden of the disorder. Out of the total global disease burden, 1% is due to epilepsy, and 80% of the epilepsy is oberved in the developing world. In some areas, 80% to 90% of people with epilepsy receive no treatment at all [1]. Analysis of data from the Quality and Outcomes Framework suggests a prevalence of 1.15% of diagnosed epilepsy in people aged 18. The prevalence of epilepsy (especially new-onset epilepsy) among the elderly population is increasing, although the possibility of misdiagnosis also remains high