ETCNet: An EEG-based motor imagery classification model combining efficient channel attention and temporal convolutional network

An average accuracy of 80.71% was achieved on the BCI Competition IV-2a dataset using an advanced end-to-end network for motor imagery signal classification.

• The brain-computer interface (BCI) uses brain signals to control external devices, particularly benefiting individuals with neuromuscular disabilities.
• Motor imagery (MI) based electroencephalogram (EEG) signals are recognized as promising for BCI applications.
• Deep learning techniques, particularly convolutional neural networks (CNN), show improved performance over traditional machine learning methods in processing MI signals.
• Challenges remain regarding subject independence and the low signal-to-noise ratio of EEG signals, complicating intention decoding.
• The proposed network integrates efficient channel attention (ECA) and temporal convolutional network (TCN) components to enhance motor imagery classification.

AI simplified