Spike Sorting and Event Detection in Neuromorphic Computing

Abstract

Neuromorphic computing is a cutting-edge field of research that focuses on designing and developing computer systems and hardware architectures inspired by the structure and functioning of the human brain. The main objective of neuromorphic computing is to create computational systems that imitate the parallel processing, learning, and energy efficiency observed in biological neural networks. Two essential tasks in the analysis of brain data from extracellular recordings are spike sorting and event detection. Spike sorting involves identifying and categorizing discrete neural spikes, while event detection focuses on recognizing specific patterns or occurrences within the spiking activity of neuromorphic networks. Typically, each neuron generates spikes with a characteristic shape, and these clusters correspond to the activity of different neurons. Here, the outcome of spike sorting is the identification of which spike corresponds to which neuron. In-depth analysis of the spike sorting and event detection which are two critical processes in interpreting the intricate neuronal activity that underpins brain function are provided. © 2024 IEEE.