Journal of Neurology & Neurophysiology

Possibility of monitoring event-related desynchronization using electroencephalography during walking

Noha Twain^{* *}Correspondence: Noha Twain, Editorial office Journal of Neurology and Neurophysiology, Brussels, Belgium,

Author info »

Introduction

Bilateral upper-limb motor imagery has been shown to be a beneficial mental activity in brain–computer interfaces based on Electroencephalography (EEG). Few researches, on the other hand, have looked at bilateral lower-limb motor imagery, and all of them have concentrated on imagining foot motions. The accuracy of left– right categorization obtained in these investigations using the EEG mu rhythm (8–13 Hz) and beta band (13–30 Hz) remained inadequate. The current study looked at the possibilities of employing lower-limb stepping motor imagery as a mental activity, as well as the EEG difference between hypothetical left-leg stepping (L-stepping) and right-leg stepping (R-stepping) motions. An experimental paradigm was devised to capture 5-second motor imagery EEG data at nine different locations around the brain's vertex. The results of eight able-bodied individuals showed that the frequently utilised mu Event-Related Desynchronisation (ERD) characteristic did not show a significant difference between the two imagined motions for all recording locations and all time intervals throughout the 5-s motor imagery period. In terms of the second often utilised aspect, beta event-related synchronisation, there was no significant difference between the two imaging tasks for the majority of the recording sites and time periods. Instead, theta band (4–8 Hz) ERD changed substantially across the L- and R-stepping imagery tasks within the first 2 s following motor imagery cue onset at five locations (FC4, C3, CP3, Cz, CPz). The current study's findings may serve as a foundation for future development of BCI systems for deciphering left and right walking during mental exercises in which the two motions are alternately visualised.

The goal of this research is to develop an efficient and feasible paradigm for brain-computer interface (BCI)-based 2-D virtual wheelchair control. The paradigm was founded on the multi-class discrimination of the spatiotemporally distinct occurrence of event-related desynchronization/ synchronization (ERD/ERS) in electroencephalogram data associated with motor execution/imagery of right/left hand movement. When compared to the traditional method of using only ERD, where bilateral ERDs appear during left/right hand mental tasks, the 2-D control demonstrated high accuracy in a short period of time, as incorporating ERS into the paradigm hypothetically enhanced the spatiotemporal feature contrast of ERS versus ERD. We also expected consumers to appreciate the simplicity of control provided by the inclusion of a noncontrol state. The control command was provided discreetly in this trial, but the virtual wheelchair moved continually. In a single visit, we examined five healthy volunteers in two sessions: motor execution and motor imagery. Each session comprised a 20-minute calibration period followed by two sets of games lasting less than 30 minutes. With motor imagery, the average target hit rate was as high as 98.4%. In the second set of wheelchair control games, every subject scored a perfect score. The average time to strike a target 10 m out was roughly 59 seconds, with the best set being 39 seconds. The higher control performance in patients who did not get substantial BCI training revealed a feasible wheelchair control paradigm for BCI users.

Several earlier investigations employing Functional Magnetic Resonance Imaging (fMRI) and Near Infrared Spectroscopy (NIRS) have revealed activations in brain regions involved in walking. These hemodynamic approaches, however, have a temporal resolution of seconds. A technology with a better temporal resolution, such as Electroencephalography (EEG), is required to detect changes within a step cycle. The problem with EEG readings while walking is that they are prone to contamination from face and neck muscles as well as movement artefacts from the electrodes and wires. Recently, several attempts to address these issues have been attempted, and brain signals associated to walking have been recorded. EEG signals might be used to infer kinematics from lower limb joints.

The event related desynchronization is the best contender for a brain signal that might be exploited in BCI in rehabilitation (ERD). This ERD is detected in the mu and beta frequency ranges during movement and movement imagining. These ERD signals have not yet been studied in the context of walking. In this study, we will look at the potential of measuring ERD while walking. In addition, the ERSPs will be examined using approaches that may be easily used in a BCI context. To separate electromyography (EMG) and EEG sources, Canonical Correlation Analysis (CCA), a blind source separation approach that offers an estimate of autocorrelation in a signal, will be employed.

Research on Brain–Computer Interface (BCIs) has been ongoing for nearly 40 years, during which time the potential of BCI applications has been investigated in a variety of domains, including communication, robotics, mobility control, and neuroprosthetics. The BCI converts intents directly into computer commands, bypassing the human body's traditional neural muscle conduction system. This allows paralysed individuals to regain independence, enhance their quality of life, and perhaps minimise the significant economic costs associated with the ongoing demands of caregivers. Our group has been working on EEG-based BCIs with the goal of producing 2-D control BCIs for possible applications such as wheelchair control. We rely on machine learning techniques to train/select effective classifiers that yield high classification accuracies using subjectspecific EEG features, as well as incorporating new physiological features to essentially improve the difference/separability between patterns generated by mental tasks, to aid in the learning process. One of the major issues of EEG-based BCI is poor decoding accuracy, which is caused by the low Signal-To-Noise Ratio (SNR) of EEG signals, which are created by the synchronised activity of millions of cortical neurons. As EEG signals are received from scalp electrodes, the spatial resolution decreases and the noise level increases. This has prompted BCI research groups to explore trustworthy features and innovative methodologies to best overcome the low SNR in order to produce a system that is both reliable and easy to use. Event-Related Desynchronization (ERD), a phenomenon of EEG amplitude attenuation during mental activities suggesting brain activity, is the most widely employed feature in cursor or wheelchair control. Users could achieve 1-D, 2-D cursor or wheelchair control by using motor imagery of hand/foot movements to detect spatially different ERD patterns, or by regulating brain rhythms to decrease signal variability and increase SNR in order to discriminate control state ERD from noncontrol state baseline activity. Recently, a 3-D virtual helicopter control game employing intelligent control techniques was described, and following a series of training sessions, a 3-D wheelchair cursor control using right/left hand and foot motor imagery was also obtained.

The Study's Purpose

To maximise level walking, prosthesis control settings are often modified. User control of prosthesis is required to modify prosthetic control settings in real time, allowing the prosthesis to adapt to changing conditions and surroundings. Smooth, uncomplicated user control of a prosthesis that matches the performance of a normal biological limb can lessen the effort and burden from the user, who will expend substantially more energy than able-bodied people in the best of conditions. This paper offers a volitional prosthesis control system based on BCIs to facilitate comfortable and effortless user control of the prosthesis, in which users can influence the prosthesis (parametric intervention) proactively by thinking alone. Automatic recognition of the user's volition, followed by automatic adjustment of prosthetic control parameters, will bring amputee gait closer to normal gait patterns, allowing the amputee to increase motion functions (e.g., upslope/downslope) and reduce energy expenditure in altered situations and environments. Meanwhile, recognising the user's conscious purpose and then controlling the prosthesis will provide the user a sense of ownership, as though "I am the one in control of prosthetic adaption." The amputee's psychological and physical well-being will improve as a result of this sense of action and control. Although this pilot research did not fully meet this goal, it is a crucial first step in achieving the aforementioned objectives. As a result, the initial phase is creating a method for the prosthetic user to alter his prosthesis in response to a basic environmental scenario with the speed and ease of thinking. Once a single switching system has been successfully implemented, the user can continue to multiple switching systems and other more sophisticated control techniques, allowing the prosthesis user to respond rapidly and efficiently to complicated environmental scenarios.