A Fully-Automated Neural Spike Sorting Based on Projection Pursuit and Gaussian Mixture Model
Kyung Hwan Kim · 2005
Existing algorithms for neural spike sorting have been unsatisfactory when the signal-to-noise ratio (SNR) is low, especially for the fully automated systems. We present a novel method that shows satisfactory performance even under low SNR, and compare its performance with the system based on principal component analysis (PCA) and fuzzy c-means (FCM) clustering algorithm. The system consists of a feature extractor that utilizes projection pursuit based on negentropy maximization, and an unsupervised classifier based on Gaussian mixture model. It is shown that the proposed feature extractor gives better performance, compared with the PCA, and the proposed combination of feature extraction and unsupervised classification yields much better performance than the PCA-FCM