A Spiking Neural Network With Adaptive Graph Convolution and LSTM for EEG-Based Brain-Computer Interfaces
Импульсная нейронная сеть с адаптивной графовой свёрткой и LSTM для интерфейсов мозг—компьютер на основе ЭЭГ
2023-01-01
SCID: 54.1/x5n8jp2h
Discuss with AI
EEG-based brain-computer interfacesadaptive graph convolutionmotor imagery decodingspike-based LSTMspiking neural networks
Figures from the paper
Abstract (AI)
Electroencephalography (EEG) signals classification is essential for the brain-computer interface (BCI). Recently, energy-efficient spiking neural networks (SNNs) have shown great potential in EEG analysis due to their ability to capture the complex dynamic properties of biological neurons while also processing stimulus information through precisely timed spike trains. However, most existing methods do not effectively mine the specific spatial topology of EEG channels and temporal dependencies of the encoded EEG spikes. Moreover, most are designed for specific BCI tasks and lack some generality. Hence, this study presents a novel SNN model with the customized spike-based adaptive graph convolution and long short-term memory (LSTM), termed SGLNet, for EEG-based BCIs. Specifically, we first adopt a learnable spike encoder to convert the raw EEG signals into spike trains. Then, we tailor the concepts of the multi-head adaptive graph convolution to SNN so that it can make good use of the intrinsic spatial topology information among distinct EEG channels. Finally, we design the spike-based LSTM units to further capture the temporal dependencies of the spikes. We evaluate our proposed model on two publicly available datasets from two representative fields of BCI, notably emotion recognition, and motor imagery decoding. The empirical evaluations demonstrate that SGLNet consistently outperforms existing state-of-the-art EEG classification algorithms. This work provides a new perspective for exploring high-performance SNNs for future BCIs with rich spatiotemporal dynamics.
Key Findings
1
Across both datasets, SGLNet consistently outperformed existing state-of-the-art EEG classification algorithms, indicating improved generality across BCI tasks.
2
SGLNet is a spiking neural network combining a learnable spike encoder, adaptive graph convolution, and spike-based LSTM for EEG-based BCIs.
3
SGLNet was evaluated on two public datasets covering emotion recognition and motor imagery decoding, representing distinct BCI applications.
4
Spike-based LSTM units capture temporal dependencies in encoded EEG spike trains, complementing the model’s spatial processing.
5
The model uses multi-head adaptive graph convolution to learn intrinsic spatial relationships among EEG channels within an SNN framework.
Research Object
EEG signals and their encoded spike trains in brain-computer interface tasks
Research Subject
spatiotemporal EEG-channel relationships and classification performance across emotion recognition and motor imagery decoding
Publication Details
Publication Date
2023-01-01
Journal
Publisher
ISSN
Cited by
116
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai4
EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces2018
Deep learning with convolutional neural networks for EEG decoding and visualization2017
Deep learning for electroencephalogram (EEG) classification tasks: a review2019
Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-Based Optimization to Spiking Neural Networks2019