Filtering by
- All Subjects: Electrical Engineering
- All Subjects: Rehabilitation
Model Based Automatic and Robust Spike Sorting for Large Volumes of Multi-channel Extracellular Data
This dissertation aims at automating the spike sorting process. A high performance, automatic and computationally efficient spike detection and clustering system, namely, the M-Sorter2 is presented. The M-Sorter2 employs the modified multiscale correlation of wavelet coefficients (MCWC) for neural spike detection. At the center of the proposed M-Sorter2 are two automatic spike clustering methods. They share a common hierarchical agglomerative modeling (HAM) model search procedure to strategically form a sequence of mixture models, and a new model selection criterion called difference of model evidence (DoME) to automatically determine the number of clusters. The M-Sorter2 employs two methods differing by how they perform clustering to infer model parameters: one uses robust variational Bayes (RVB) and the other uses robust Expectation-Maximization (REM) for Student’s 𝑡-mixture modeling. The M-Sorter2 is thus a significantly improved approach to sorting as an automatic procedure.
M-Sorter2 was evaluated and benchmarked with popular algorithms using simulated, artificial and real data with truth that are openly available to researchers. Simulated datasets with known statistical distributions were first used to illustrate how the clustering algorithms, namely REMHAM and RVBHAM, provide robust clustering results under commonly experienced performance degrading conditions, such as random initialization of parameters, high dimensionality of data, low signal-to-noise ratio (SNR), ambiguous clusters, and asymmetry in cluster sizes. For the artificial dataset from single-channel recordings, the proposed sorter outperformed Wave_Clus, Plexon’s Offline Sorter and Klusta in most of the comparison cases. For the real dataset from multi-channel electrodes, tetrodes and polytrodes, the proposed sorter outperformed all comparison algorithms in terms of false positive and false negative rates. The software package presented in this dissertation is available for open access.
Neuromodulation is an emerging field of research that has a proven therapeutic benefit on a number of neurological disorders, including epilepsy and stroke. It is characterized by using exogenous stimulation to modify neural activity. Prior studies have shown the positive effect of non-invasive trigeminal nerve stimulation (TNS) on motor learning. However, few studies have explored the effect of this specific neuromodulatory method on the underlying physiological processes, including heart rate variability (HRV), facial skin temperatures, skin conductance level, and respiratory rate. Here we present preliminary results of the effects of 3kHz supraorbital TNS on HRV using non-linear (Poincaré plot descriptors) and time-domain (SDNN) measures of analysis. Twenty-one (21) healthy adult subjects were randomly assigned to 2 groups: 3kHz Active stimulation (n=11) and Sham (n=10). Participants’ physiological markers were monitored continuously across three blocks: one ten-minute baseline block, one twenty-minute treatment block, and one ten-minute recovery block. TNS targeting the ophthalmic branches of the trigeminal nerve was delivered during the treatment block for twenty minutes in 30 sec. ON/OFF cycles. The active stimulation group exhibited larger values of all Poincaré descriptors and SDNN during blocks two and three, signifying increased HRV and autonomic nervous system activity.