Dual stream neural networks for brain signal classification

摘要

Objective. The primary objective of this work is to develop a neural nework classifier for arbitrarycollections of functional neuroimaging signals to be used in brain–computer interfaces (BCIs).Approach. We propose a dual stream neural network (DSNN) for the classification problem. Thefirst stream is an end-to-end classifier taking raw time-dependent signals as input and generatingfeature identification signatures from them. The second stream enhances the identified featuresfrom the first stream by adjoining a dynamic functional connectivity matrix aimed atincorporating nuanced multi-channel information during specified BCI tasks. Main results. Theproposed DSNN classifier is benchmarked against three publicly available datasets, where theclassifier demonstrates performance comparable to, or better than the state-of-art in each instance.An information theoretic examination of the trained network is also performed, utilizing varioustools, to demonstrate how to glean interpretive insight into how the hidden layers of the networkparse the underlying biological signals. Significance. The resulting DSNN is a subject-independentclassifier that works for any collection of 1D functional neuroimaging signals, with the option ofintegrating domain specific information in the design.