Spike Detection and Clustering With Unsupervised Wavelet Optimization in Extracellular Neural Recordings

Abstract

Automatic and accurate detection of action potentials of unknown waveforms in noisy extracellular neural recordings is an important requirement for developing brain-computer interfaces. This study introduces a new, wavelet-based manifestation variable that combines the wavelet shrinkage denoising with multiscale edge detection for robustly detecting and finding the occurrence time of action potentials in noisy signals. To further improve the detection performance by eliminating the dependence of the method to the choice of the mother wavelet, we propose an unsupervised optimization for best basis selection. Moreover, another unsupervised criterion based on a correlation similarity measure was defined to update the wavelet selection during the clustering to improve the spike sorting performance. The proposed method was compared to several previously proposed methods by using a wide range of realistic simulated data as well as selected experimental recordings of intracortical signals from freely moving rats. The detection performance of the proposed method substantially surpassed previous methods for all signals tested. Moreover, updating the wavelet selection for the clustering task was shown to improve the classification performance with respect to maintaining the same wavelet as for the detection stage.

A Wireless Headstage for Combined Optogenetics and Multichannel Electrophysiological Recording

“…2) Other Thresholding Method: The proposed algorithm is compared with the median of the absolute deviation (MAD) operator, which is employed in common spike sorting software, such as Waveclus an offline spike sorting freeware [29]. The MAD operator is computed as follows: σ = median{|x|/0.6745}, where x is the band-pass filtered input signal.…”

Automatic spike sorting by unsupervised clustering with diffusion maps and silhouettes

“…Alternatively, an algorithm for automatic unsupervised detection of action potentials in extracellular recordings is introduced in Shalchyan et al [11]. A new manifestation variable for detection is defined based on the combination of denoised wavelet coefficients over selected scales.…”

Automatic extracellular spike denoising using wavelet neighbor coefficients and level dependency

“…The wavelet transform is a timefrequency decomposition ideally suited for non-stationary signals, which has been widely used in the field of neural signals processing. Wavelet-based denoising techniques have been shown to be valuable tools for spike denoising [5][6][7][8]. Common to all these methods is that a threshold is applied in wavelet domain to keep a few large coefficients that contribute the signals as originally proposed by Donoho [9].…”