Skip to main content
Fig. 8 | Brain Informatics

Fig. 8

From: Hemodynamic functional connectivity optimization of frequency EEG microstates enables attention LSTM framework to classify distinct temporal cortical communications of different cognitive tasks

Fig. 8

Comparison of performance of the deep learning frameworks. A Accuracy, B precision, C recall. Each architecture was independently optimized with 4, 6, 8, 10, 12, 14, 15 and 16 LSTM input cells for temporal EEG sampling of 750, 500, 375, 300, 250, 214, 200 to 180 ms/segment, respectively

Back to article page