A Review of EEG-Based Computer-Aided Detection of Brain Disorders Using a Different Strategy

1. Introduction

The brain governs all tasks performed by the human body and analyses information from the environment. The brain governs various activities as intelligence, emotion, creativity, and memory. The brain collectively processes the information from the senses to make the decision and store it for further analysis. The brain regulates our thoughts, memory, speech, and the operation of the organs in our body. The brain and spinal cord are connected by the brainstem and parts of the Central Nervous System (CNS). The cerebral hemispheres are separated into lobes by fissures. There are four lobes in each hemisphere: frontal region, temporal region, parietal region, and occipital region. Each hemisphere lobe can be further categorized into areas that perform distinct duties. The interactions between the cerebral lobes and the right and left hemispheres are quite intricate. In the brain, There are two different types of cells in the brain: glia cells and neurons, which are nerve cells. Neurons have dendrites, a cell body, and an axon. Electrical and chemical impulses are used by nerve cells to deliver and receive information. An activated neuron will communicate its energy to all other neurons in its surroundings. Neurons communicate with one another by transmitting energy across a gap called a synapse. A neuron contains multiple fibers known as dendrites, taking up information from neighboring neurons. The information from dendrites and send to the cell body and it is decided for further processing. Brain disorders are caused by the impairment of the different brain regions. This chapter provides a comprehensive exploration of using EEG as a foundational signal for the automated diagnosis of various brain disorders, including epilepsy, autism, Alzheimer's disease, Parkinson's disease, schizophrenia, and ischemic brain stroke.

EEG signals are nonstationary, nonlinear, and non-Gaussian in nature. The acquisition of EEG signals is a non-invasive technique and is widely used for diagnosing brain disorders. Numerous conditions, such as brain tumors, problems associated with trauma, epilepsy, depression, and cerebrovascular lesions can be diagnosed using EEG signals. EEG traces are dissimilar for different responses of the brain. Generally, various signal processing techniques are applied to EEG signals to distinguish between abnormal and normal behaviour of the brain. These signals are produced by electrodes that are placed at various points on the scalp. EEG recordings are divided into two types: (i) monopolar (ii) bipolar. The voltage difference between an active electrode that is mounted on the scalp and a reference electrode that is mounted on the ear lobe gives us monopolar EEG recordings, whereas the voltage difference between two (active and reference) electrodes placed on the scalp alone gives us bipolar recordings. The position of electrodes placed on the scalp is always done according to the International 10-20 Standard for the placement of electrodes, i.e., Fp1, Fp2, F3, F4, T5, T6, F7, F8, Fz, Cz, Pz, O1, O2, T3, C3, C4, T4, P3, P4. Figure 1 depicts the Standard.

Figure 1.

10-20 international standard for the placement of the EEG electrodes.

EEG signals can be categorized into five different rhythms or bands based on their frequencies: delta, theta, alpha, beta, and gamma. The delta band has a frequency range of 1-4 Hertz and is primarily associated with the depth of sleep, with higher delta rhythm indicating a deeper sleep. Additionally, an increase in delta waves is linked to an increase in internal working memory tasks such as concentration. The theta band, on the other hand, has a frequency range of 4-7 Hertz and is linked to cognitive workload and memory encoding and retrieval. The alpha band ranges from 7-12 Hertz and is prominent when subjects close their eyes and enter a peaceful state. These waves are used in biofeedback training to examine a person's calm state and are sometimes associated with attention and inhibition. The beta band has a frequency range of 12-30 Hertz and is observed during motor functions, with stronger beta wave frequencies indicating more movement in the body. There is evidence to suggest that the human brain may mimic limb movements observed in others, possibly synchronized by the beta band. Finally, the gamma band has a frequency greater than 30 Hertz, typically 40 Hertz, and serves as a carrier frequency to exchange data between different parts of the brain and reflect attentive focus. Gamma waves are also associated with rapid eye movements or micro-saccades.