Evaluation of Spike-Detection Algorithms for a Brain-Machine Interface Application

Abstract

Real time spike detection is an important requirement for developing brain machine interfaces (BMIs). We examined three classes of spike-detection algorithms to determine which is best suited for a wireless BMI with a limited transmission bandwidth and computational capabilities. The algorithms were analyzed by tabulating true and false detections when applied to a set of realistic artificial neural signals with known spike times and varying signal to noise ratios. A design-specific cost function was developed to score the relative merits of each detector; correct detections increased the score, while false detections and computational burden reduced it. Test signals both with and without overlapping action potentials were considered. We also investigated the utility of rejecting spikes that violate a minimum refractory period by occurring within a fixed time window after the preceding threshold crossing. Our results indicate that the cost-function scores for the absolute value operator were comparable to those for more elaborate nonlinear energy operator based detectors. The absolute value operator scores were enhanced when the refractory period check was used. Matched-filter-based detectors scored poorly due to their relatively large computational requirements that would be difficult to implement in a real-time system.

A unified framework and method for automatic neural spike identification

“…1: (1) detection of temporal segments of the voltage trace that are likely to contain spikes, (2) estimation of a set of features for each segment, and (3) classification of the segments according to these features. A variety of methods exist for solving each step (e.g., (1) thresholding based on absolute value (Obeid and Wolf, 2004), squared values (Rutishauser et al, 2006), Teager energy (Choi et al, 2006), or other nonlinear operators (Rebrik et al, 1999), (2) features such as peak-to-peak width/amplitude, projections onto principal components (Lewicki, 1998), or wavelet coefficients (Quiroga et al, 2004;Kwon and Oweiss, 2011), and (3) classification methods such as K-means (Lewicki, 1998), mixture models (Sahani, 1999;Shoham et al, 2003), or superparamagnetic methods (Quiroga et al, 2004)). …”

Real-time signal detection and classification algorithms for body-centered systems

“…Given the importance of spike detection, it is interesting to be able to quantitatively assess the quality of any implemented detector [20], [179].…”

“…The SC stage is based on spike detection [20] yielding an acceptable compromise between signal accuracy and process load. Although quite popular in literature (see [20] and references therein), we have included a novel noise-tracking algorithm for adaptive threshold setting that outperforms the published basic algorithms.…”

“…Although quite popular in literature (see [20] and references therein), we have included a novel noise-tracking algorithm for adaptive threshold setting that outperforms the published basic algorithms. The RM stage implements a stepped reduction in the transmitted data rate as the channel bandwidth decreases, enforcing its simplicity for sake of process load saving.…”

“…As it has widely been discussed in the literature [20], quantitatively assessing spike detection requires knowledge of the ground truth. Recordings from micro-electrode arrays do not allow intracellular recording which means that the ground truth is not known.…”