Improvement of Nociceptive Spike Clusterization with Shape Approximation

Abstract

© 2017, Springer Science+Business Media, LLC. Cluster spike analysis is widely used for studies of neuronal activity when electrical signals are sorted out and grouped according to the different shapes. We recently applied this method to sort out the nociceptive spikes in the trigeminal nerve implicated in generation of migraine pain. However, the electrical noise leading to less accuracy of calculated spike parameters often hinder the correct sorting of nerve signals. In this study, in order to improve the accuracy of calculations, we explored the prior approximation of spike shapes before applying clusterization. The prior fitting of spike shapes allowed us to extract signal parameters much more precisely and detect the strongly increased number of spike clusters which is close to the expected number of fibers in the trigeminal nerve. Prior approximation improved cluster analysis outcomes and, importantly, revealed new clusters that demonstrated the different functional properties, suggesting that their function was previously hidden within the multiple firing.