4D-CRNN Model of EEG Motor Imagery with Attention Module

EEG signals are often used to classify motor imagery in the field of brain-computer interface research. Deep learning is used in EEG signal classification to focus on extracting features, selecting features, and classifying them. Taking the frequency, spatial and temporal features of the signal into account at the same time, in the 4D-CRNN model, motor imagery can be classified from EEG signals based on its characteristics. The model achieved good results in this area. As a part of the experiment, attention module has been introduced into the model, resulting in a partially improved model effect.