Convolutional Spiking Neural Networks for Detecting Anticipatory Brain Potentials Using Electroencephalogram

Abstract

Spiking neural networks (SNNs) are receiving increased attention as a means to develop 'biologically plausible' machine learning models. These networks mimic synaptic connections in the human brain and produce spike trains, which can be approximated by binary values, precluding high computational cost with floating-point arithmetic circuits. Recently, the addition of convolutional layers to combine the feature extraction power of convolutional networks with the computational efficiency of SNNs has been introduced. In this paper, the feasibility of using a convolutional spiking neural network (CSNN) as a classifier to detect anticipatory slow cortical potentials related to braking intention in human participants using an electroencephalogram (EEG) was studied. The EEG data was collected during an experiment wherein participants operated a remote controlled vehicle on a testbed designed to simulate an urban environment. Participants were alerted to an incoming braking event via an audio countdown to elicit anticipatory potentials that were then measured using an EEG. The CSNN's performance was compared to a standard convolutional neural network (CNN) and three graph neural networks (GNNs) via 10-fold cross-validation. The results showed that the CSNN outperformed the other neural networks, and had an average predictive accuracy of 99.06% with an average true positive rate (TPR) of 98.50% and an average true negative rate (TNR) of 99.20%. Two-sample t-tests showed that the higher performance of the CSNN was not significantly different compared to the CNN, but significant compared to the GNNs. Classification performance degradation as a result of converting the floating-point EEG data into spike trains via delta modulation was also studied for the CSNN. The best threshold value was found to be 0.5, achieving an average accuracy of 97.56% with an average TPR of 91.64% and an average TNR of 99.03%. Two-sample t-tests further showed that statistically similar performance could be obtained even when the threshold was varied from 0.375 to 0.625.