Multineuronal spike classification based on multisite electrode recording,whole-waveform analysis, and hierarchical clustering

We proposed here a method of multineuronal spike classification based on mu ltisite electrode recording, whole-waveform analysis, and hierarchical clus tering for studying correlated activities of adjacent neurons in nervous sy stems. Multineuronal spikes were recorded with a multisite electrode placed in the hippocampal pyramidal cell layer of anesthetized rats, If the imped ance of each electrode site is relatively low and the distance between elec trode sites is sufficiently small, a spike generated by a neuron is simulta neously recorded at multielectrode sites with different amplitudes. The cov ariance between the spike waveform at each electrode site and a template wa s calculated as a damping factor due to the volume conduction of the spike from the neuron to the electrode site. Calculated damping factors were vect orized and analyzed by hierarchical clustering using a multidimensional sta tistical test. Since a cluster of damping vectors was shown to correspond t o an antidromically identified neuron, spikes of different neurons are clas sified by referring to the distributions of damping vectors. Errors in damp ing vector calculation due to partially overlapping spikes mere minimized b y successively subtracting preceding spikes from raw data. Clustering error s due to complex spike bursts (i.e., spikes with variable amplitudes) were avoided by detecting such bursts and then using only the first spike of a b urst for clustering, These special procedures produced better cluster separ ation than conventional methods, and enabled multiple neuronal spikes to be classified automatically. Waveforms of classified spikes were well superim posed, We concluded that this method is particularly useful for separating the activities of adjacent neurons that fire partially overlapping spikes a nd/or complex spike bursts.