Studies on Electroencephalogram for Upper Limb Rehabilitation

Introduction

The upper limb rehabilitation from Electroencephalogram (EEG) signals is decoded to improve the quality of lives of individuals with motor impairments and provide alternative pathways for interacting with their environment. Upper limb functionality is critical for performing daily activities such as eating, drinking, writing, typing, and grooming, as well as for engaging in sports and leisure activities. The human brain's electrical activity can be captured non-invasively using a device to capture the signals produced by the neurons (Chakraborty et al., 2020). EEG is a safe, painless, and valuable method in clinical and research settings to study brain function and diagnose conditions such as stress, epilepsy, and sleep disorders (Ofner et al., 2017).

During the data acquisition, the electrodes are usually placed with a conductive gel to improve signal quality (Das et al., 2020). These electrodes are connected to an amplifier that amplifies the electrical signals produced by the neurons in the brain. The EEG signal reflects the activity of millions of neurons firing in synchrony. These signals are measured in the form of waves with different frequencies, which are referred to as brain waves. The different lobes of human brain are shown in Figure 1, which illustrates the various parts of the brain. Every lobe is responsible for different functions, making up the human brain.

Figure 1.

Representation of the distinct anatomical sections of the human brain

EEG-based motor imagery (MI) signal analysis has potential applications in rehabilitation and assistive technology, enabling individuals with disabilities to control robots or virtual devices using their thoughts. This involves understanding the joints and movements of the upper limb and translating them into code that the robotic manipulator can execute. The upper limb consists of different joints, like the shoulder, elbow, wrist, and hand, each responsible for various directional and angular movements to perform daily activities (Zhang et al., 2019). To efficiently program a manipulator to mimic these movements, we must consider its degrees of freedom, each joint, and its range of motion.

For example, the shoulder joint can move in several directions, including forward, backward, sideways, and rotate. The elbow joint can bend and straighten, while the wrist joint can flex, extend, and rotate. The fingers and thumb can also move independently and together to grasp objects. To program a robotic arm to perform these movements, we must use an optimal set of sensors and actuators to detect and respond to environmental changes. This can involve using a combination of sensors, such as position, force, and tactile sensors, to detect the position and force of the arm and the objects it interacts with. When programmed, the robotic arm mimics the movements of the upper limb; we can use it to perform a wide range of tasks, from simple pick-and-place operations to more complex manipulations and interactions with the environment.

In the following section of this chapter, we will discuss various stages of signal preprocessing, different types of upper limb movements and the methods used to classify based on features and represent these movements for purposes (pictorially given in Figure 2). Also, the techniques used in EEG signals are discussed in the current study. The chapter provides the strengths and limitations of each technique and how they can be applied in different scenarios and explores the various applications of movements in rehabilitation and motor control, human-robot interaction, gaming, and virtual reality. It highlights the potential benefits of using these applications and how they can improve the lives of individuals with motor damages.

Figure 2.

Working process of electroencephalogram signal