International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 07 | July 2023
p-ISSN: 2395-0072
www.irjet.net
EEG Signal Classification using Deep Neural Network Shakir Ali1, Mohammad Shahrookh Husain2, Pankaj Mishra3 1Assistant Professor for the CSE Department, ACE Engineering College, Hyderabad 2 Research Scholar for Jamia Millia Islamia Central University, New Delhi 3Assistant Professor for the CSE Department, ACE Engineering College, Hyderabad
----------------------------------------------------------------------------***-------------------------------------------------------------------------processed. To build the feature mode of recurrence Abstract: - Electroencephalography (EEG) is a non-
plot, Features in the time- and frequency-domains are extracted, respectively.
invasive technique used to measure and record the electrical activity of the brain. It plays a crucial role in various medical and research applications, including brain-computer interfaces, neurological disorder diagnosis, and cognitive state monitoring. In recent years, deep learning has emerged as a powerful tool for extracting meaningful patterns from EEG signals, leading to significant advancements in EEG signal classification tasks. This research focuses on the application of deep neural networks (DNNs) for EEG signal classification. The primary objective is to develop a robust and accurate classification model capable of identifying distinct brain states or patterns associated with specific mental activities or neurological conditions. To achieve this, we propose a multi-layered deep neural network architecture, leveraging convolutional layers to automatically learn spatial features and recurrent layers to capture temporal dependencies in the EEG data. Neural networks find utility in a several uses because the combination of classification being available in deep learning approaches. This study employs SVM to categorize EEG signals. recording of the brain's electrical impulses known as an electroencephalogram (EEG) can be used to diagnose many medical disorders. Parts of the brain are affected by partial epilepsy, and the EEG recorded from those areas is known as FocalEEG, whereas the EEG recorded from another area is referred to along with Non-Focal EEG. When a patient has drug-resistant epilepsy, the Focal EEG identification helps the doctors locate the epileptogenic focus and, as a result, recommend surgical removal of those brain regions. In this, we have suggested a methodology for categorizing nonfocal and focal EEG. Recent years have seen a growth in the utilization of a brain-computer interface has made it potential to investigate the brain's control mechanism using EEG signal. An effective focus based on a classification technique Recurrence plot CNN are suggested the address to issue of EEG signal categorization. To increase the signal intensity during the workout interval, EEG signals are first pre-
© 2023, IRJET
|
Impact Factor value: 8.226
Keyword: EEG, Recurrence Plot, Support Vector Machine (SVM), Neural Network (NN).
I. INTRODUCTION EEG signal which is used to detect brain related disease. A cheap instrument for analyzing the brain activity captured with some electrodes on the scalp is the EEG signal. Neurology specialists examine the signal visually to ascertain the beginning of epilepsy. Nonetheless, properly analyzing EEG signals is laborious, time - consuming, frequently results in incorrect alarms detection of epilepsy. To overcome the problem, epilepsy can be automatically detected using EEG signals considered. In cases of conditions, a person with epilepsy may experience sudden seizures that cause his muscles to twitch and even cause him to lose consciousness. EEG is a monitoring of "electrical" activity in the brain made from the scalp. The frequencies that were captured demonstrate the "electrical" behavior of the brain. A person with epilepsy suffers from sudden seizures that cause convulsions in their muscles and, in some cases, even cause them to lose consciousness. 50 million individuals worldwide, or 1% of the population, suffer from epilepsy. A person with this disease is not well regarded in society, even to the point where marriages are forbidden. Although the sickness itself does not cause a harmful state, a loss of consciousness can be hazardous if the person is driving or swimming. The electroencephalogram (EEG) is frequently used to examine different aspects of brain function. One such condition where EEG is used for clinical analysis is epilepsy. Some epileptics develop treatment resistance and require surgical excision of the brain regions responsible for the condition to regain health. The epileptogenic foci are those areas of the brain that result in epileptic seizures. The frequency of these operations has successfully
|
ISO 9001:2008 Certified Journal
|
Page 1150