A Novel Classification Framework Using the Graph Representations of Electroencephalogram for Motor Imagery Based Brain-Computer Interface

Abstract

The motor imagery (MI) based brain-computer interfaces (BCIs) have been proposed as a potential physical rehabilitation technology. However, the low classification accuracy achievable with MI tasks is still a challenge when building effective BCI systems. We propose a novel MI classification model based on measurement of functional connectivity between brain regions and graph theory. Specifically, motifs describing local network structures in the brain are extracted from functional connectivity graphs. A graph embedding model called Ego-CNNs is then used to build a classifier, which can convert the graph from a structural representation to a fixed-dimensional vector for detecting critical structure in the graph.We validate our proposed method on four datasets, and the results show that our proposed method produces high classification accuracies in two-class classification tasks (92.8% for dataset 1, 93.4% for dataset 2, 96.5% for dataset 3, and 80.2% for dataset 4) and multiclass classification tasks (90.33% for dataset 1). Our proposed method achieves a mean Kappa value of 0.88 across nine participants, which is superior to other methods we compared it to. These results indicate that there is a local structural difference in functional connectivity graphs extracted under different motor imagery tasks. Our proposed method has great potential for motor imagery classification in future studies.

A Novel Deep Learning Method Based on an Overlapping Time Window Strategy for Brain–Computer Interface-Based Stroke Rehabilitation

“…The aim of graph neural networks (GNNs) is to use graph structure data and node features as input to learn a representation of the node (or graph) for relevant tasks [ 23 ]. Because EEG data are easily converted to graph structure data, several studies have investigated GNNs applied to EEG signal-based tasks [ 17 , 24 , 25 , 26 , 27 ]. An important aspect of using a GNN to classify EEG signals is building graph data, the original data first need to be converted into graph structure data.…”

“…Lin et al developed a convolutional neural network (CNN)-based model for predicting BCI rehabilitation outcomes [ 15 ]. Liang et al used the long short-term memory (LSTM) neural network for generating motor trajectories of the lower-extremity exoskeleton for stroke rehabilitation [ 16 ] and a graph embedding-based model, Ego-CNN, for identifying key graph structures during MI [ 17 ]. However, BCI systems using DL methods require large amounts of EEG data for training models, which results in a bottleneck in therapy.…”

Learning Spatiotemporal Graph Representations for Visual Perception Using EEG Signals

“…They obtained 84.35% as classification accuracy for three classes. Similarly, Jin et al [39] used PLV values and graph representations to classify motor imagery tasks. Another work used the Pearson correlation coefficient for the graph construction [40].…”

Motor Imagery EEG Classification Based on Riemannian Sparse Optimization and Dempster-Shafer Fusion of Multi-Time-Frequency Patterns

“…Motor imagery (MI) is the mental imagination of body movement without actual muscle movement and its corresponding rhythmic activities of the brain could be observed and applied as the input signals of BCI systems [4]. The regular rhythmic power changes in the sensorimotor area within mu (8)(9)(10)(11)(12) and beta (13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) frequency bands are called event-related desynchronization/ synchronization (ERD/ERS), which can be detected to discriminate different kinds of MI tasks [5]. MI-based BCIs are flexible in their applications for consideration of MI not demanding external stimuli [6].…”

“…Whereas, the effectiveness of CSP is very sensitive to the selection of frequency bands and easily affected by irrelevant noise [16]. Thus, a fixed time window for feature extraction may lead to low classification accuracy [17]. There are numerous extensions of traditional CSP to optimize subjectspecific operational time windows or frequency bands, such as sub-band CSP (SBCSP) [18], filter bank CSP (FBCSP) [19] and sliding window discriminative CSP (SWDCSP) [20].…”