Motor imagery

About: Motor imagery is a research topic. Over the lifetime, 4158 publications have been published within this topic receiving 126962 citations.

Recognizing Motor Imagery Tasks Using Deep Multi-Layer Perceptrons

Abstract: A brain-computer interface provides individuals with a way to control a computer. However, most of these interfaces remain mostly utilized in research laboratories due to the absence of certainty and accuracy in the proposed systems. In this work, we acquired our own dataset from seven able-bodied subjects and used Deep Multi-Layer Perceptrons to classify motor imagery encephalography signals into binary (Rest vs Imagined and Left vs Right) and ternary classes (Rest vs Left vs Right). These Deep Multi-Layer Perceptrons were fed with power spectral features computed with the Welch’s averaged modified periodogram method. The proposed architectures outperformed the accuracy achieved by the state-of-the-art for classifying motor imagery bioelectrical brain signals obtaining 88.03%, 85.92% and 79.82%, respectively, and an enhancement of 11.68% on average over the commonly used Support Vector Machines.

An Improved Approach for Extracting Features and Classifying Motor Imagery EEG Signals Through Machine Learning

Abstract: The brain–computer interface allows interaction between human brain and a machine. BCI is mainly helpful for disabled persons to do their day-to-day activities independently. Brain can communicate with the machine through EEG (non-invasive method) from user’s brain cortex. This paper focuses on controlling the hand and foot movements based on human’s thoughts. The EEG recordings are taken from datasets of BCI competition III. The electroencephalogram (EEG) signal features are extracted using common spatial pattern combined with power spectral density and Hilbert transform. The classification of signals based on right hand and right foot movement is done by linear discriminant analysis (LDA) method. This proposed method is tested on five healthy subjects, and average accuracy is improved to 85.76%. In future, the classified output is converted into command signals and can able to control the robotic devices such as arm or wheelchair which will assist the people with motor impairment.

Using Motor Imagery Training to Improve Reach Estimation in the Elderly: Implications for Reducing Fall Risk

Abstract: According to recent Center for Disease Control and Prevention statistics, falls are one of the main causes of accidents and fatalities among the elderly. Whereas a several factors may contribute to falls, it has been suggested that weak mental representation of intended actions is a factor. For example, in a reach setting, many older adult’s overor underestimation reach abilities, thus posing a higher risk due to loss of postural control. The intent of this study was to determine if a reach-specific motor imagery training program could improve reach planning and potentially reduce fall risk. The present study involved a group of 23 older adult participants, aged 65-81 years, divided into three groups: a control group and two intervention groups categorized by age, 65to 73 years and 74to 81 years. Intervention groups were administered a reach-specific imagery training program three days a week over the course of 4 weeks. Participants were preand post-tested on estimation of reach via use of motor imagery in three conditions: seated, standing-on-2-feet, and standing-on-1-foot. Results indicated that both intervention groups significantly improved their reach estimation, p<.05, whereas the control group scores did not differ. No noticeable difference was seen between the two intervention groups or between reach conditions. These findings suggest that motor imagery training has promise as ineffective tool in reducing fall risk among the elderly.

Multi-modal feedback for bci based stroke rehabilitation: a case study

Abstract: In recent years, a variety of different BCI applications for communication and control were developed. A promising new idea is to utilize BCI systems as tools for brain rehabilitation. The BCI can detect the user's movement intention and provide online feedback for rehabilitation sessions. Both, functional electrical stimulation (FES) systems and brain-computer interface (BCI) based rehabilitation are earning year by year more involvement within the rehabilitation field. This paper presents the coupling of a motor imagery based BCI system with two feedback modalities. A multichannel neurostimulator is controlling both hands, performing an extension of the fingers. Secondly, a first- person Virtual Reality Avatar performs the same motor movements. The effectiveness of the proposed method has been tested on a 65 year old stroke patient, performing fourteen rehabilitation sessions within six weeks.

Discrimination of Motor Imagery from Functional MR Image of Human Brain Using Multi-ROI 3D Convolutional Neural Networks

Abstract: Motor imagery (MI), a covert cognitive process where an individual mentally simulate an action without actually moving any body part, can provide an effective neurorehabilitation tool for motor function improvement or recovery. MI can become more efficient by providing feedback to the patient indicating whether he/she employs MI correctly or not. In this study, we applied 3D convolutional neural networks (3D-CNNs) to brain images acquired using functional MRI to classify the brain state. As the result, It was suggested that MI classification using our 3D-CNNs is possible and the classifier works for unknown subjects’ data.