Automated spike sorting algorithm based on Laplacian eigenmaps and k-means clustering.

Take-home message: spike-sorting is a difficult and time-consuming process involving a combination of operator judgement and automated processing. Here, we propose a robust method of automatic spike sorting based on the combination of Laplacian eigenmaps (LE) feature extraction and the k-means clustering algorithm. The performance of the proposed method is compared with systems based on simple amplitude features and on PCA-derived features.… Read More