Automated Seizure Classification Using Deep Neural Network Based on Autoencoder

Automated Seizure Classification Using Deep Neural Network Based on Autoencoder

Rahul Sharma, Pradip Sircar, Ram Bilas Pachori
DOI: 10.4018/978-1-7998-2120-5.ch001
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

A neurological abnormality in the brain that manifests as a seizure is the prime risk of epilepsy. The earlier and accurate detection of the epileptic seizure is the foremost task for the diagnosis of epilepsy. In this chapter, a nonlinear deep neural network is used for seizure classification. The proposed network is based on the autoencoder that significantly explores the non-linear dynamics of the electroencephalogram (EEG) signals. It involves the traditional deep neural domain expertise to extract the features from the raw data in order to fit a deep neural network-based learning model and predicts the class of the unknown seizures. The EEG signals are subjected to an autoencoder-based neural network that unintendedly extracts the significant attributes that are applied to the softmax classifier. The achieved classification accuracy is up to 100% on different publicly available Bonn University database classes. The proposed algorithm is suitable for real-time implementation.
Chapter Preview
Top

Introduction

The brain is the central command center that controls most of the human activities. It consists of billions of neurons that play a vital role in information exchange, which takes place through electrical and chemical forms. The neural activities can be detected with the electroencephalogram (EEG) signals. The EEG signals are generated through the excitatory and inhibitory postsynaptic potentials of neurons (Jirsa et al, 2014). The EEG signals are particularly useful for evaluating patients with suspected seizures, epilepsy, and unusual neurological problems. It is highly sensitive to show sudden changes in neural functioning even as they first occur (Bassett, D.S., & Sporns, O., 2017). The primary purpose of the recorded EEG signals is to evaluate patients with known seizures that can permit accurate diagnosis of the seizures type and epilepsy syndrome so that therapy may be appropriately directed or to diagnose unknown symptoms that may represent seizures. The seizures classification is yet a prime task, as a seizure affected person may continue to show normal brain activities, while the abnormalities in brain activity may occur due to head injury or high fever, which is not always a sign of a seizure (Pati, S., & Alexopoulos, A.V., 2010). Hence, the type of seizures depends on the origin of a disturbance inside the brain and how far it spreads.

The recurrent inexcusable seizures are known as epilepsy, and these types of seizures cannot be controlled with medication. The neurological abnormality in the brain manifests as a seizure that is the prime risk of epilepsy. The earlier and accurate detection of epileptic seizures is the foremost task for the diagnosis of epilepsy. According to the world health organization report, nearly 50 million persons are affected by epilepsy every year, and ninety percent of total epileptic humans belong to developing countries (WHO, 2015). The causes may be a lack of minerals, vitamins, medicine, knowledge, etc. Almost 60-70% of seizures are automatically cured with age, proper medication, and meditation, while the remaining are only cured with brain surgery. In surgery, the doctors remove the infected brain area by the operation. Before this process, the person passes through various observations intensively to detect the location in the brain from where the seizures begin.

Complete Chapter List

Search this Book:
Reset