Fast Template Matching for Spike Sorting

The present study introduces an approach to detect and classify extracellularly recorded action potentials of neurons, usually termed as spike sorting. Our approach is based on template matching, which is an optimal filter under Gaussian noise; however, this approach is usually expensive in computational time.To speed up the filter, it is important to curtail the matching process when the distance between template and waveform exceeds some threshold. We approach this aspect of the problem using the frame of similarity detection algorithms (SSDA) and Davies-Bouldin validation indices (DBVI). Windowing pair of the filter was selected in DBVI based order and a signal which has rapidly increasing error was discarded to reduce the computational time. DBVI is a function of the ratio of the sum of within-cluster scatter to between-cluster separation, thus using this order we can expect to separate a signal and a noise in fewer window point calculation than full point matching. This matching process performed well, with a shorter computational time and fewer incorrect classifications than other ordering methods such as time based or amplitude based order.