Fig. 3From: Hemodynamic functional connectivity optimization of frequency EEG microstates enables attention LSTM framework to classify distinct temporal cortical communications of different cognitive tasksSchematic diagram of Attention-based LSTM deep learning framework adopted in the study. 'T' represents the temporal, sliding input sampling window size and is chosen between 180 to 750Â ms/segment associated with the specific number of cells ranging from 4 to 16, based on the stability of transition probabilities of the optimized microstatesBack to article page