Spike sorting based on shape, phase, and distribution features, and K-TOPS clustering with validity and error indices

Area of Science:

• Neurophysiology
• Computational Neuroscience
• Data Analysis

Background:

• Spike sorting is crucial for neurophysiology, impacting subsequent analyses.
• Feature extraction and spike clustering are critical steps in spike sorting.

Purpose of the Study:

• To develop and validate a new spike sorting approach.
• To improve the accuracy of classifying neural events and assessing neuron properties.

Main Methods:

• A novel spike sorting approach using shape, phase, and distribution features (SS-SPDF).
• An efficient clustering algorithm (K-TOPS) with K-means and template optimization in phase space.
• Integration of validity and error indices for clustering verification.

Main Results:

• The SS-SPDF and K-TOPS method demonstrated efficient sorting of single-unit and overlapping spikes.
• The approach showed superior performance in classifying spikes and neurons compared to existing methods.
• Effective assessment of neuron modulating properties was achieved using the new method.

Conclusions:

• The proposed spike sorting method offers an efficient and accurate solution for neural data analysis.
• This approach enhances the classification of neural events and the understanding of neuronal activity.
• The method is effective for analyzing extracellular recordings in complex behavioral tasks.