Spike Sorting based on Improved Fuzzy C-means
LIU Han;LI Zhenxin;YU Yi;DONG Bingchao;School of Biomedical Engineering,Xinxiang Medical University;
To propose an improved fuzzy clustering algorithm combined with fuzzy clustering algorithm and subtractive clustering method( SCM) to classify the detected spikes. The number and position of the clustering centers were obtained quickly by SCM,and then the obtained clustering centers as initial values were applied to the fuzzy clustering algorithm. The results indicated that this algorithm could reduce the dependence on initial clustering centers of fuzzy clustering algorithm,save the iterative computation time of fuzzy clustering algorithm and improve the operation efficiency. At the same time,this algorithm improved the classification accuracy of the spikes,especially with the SNR being less than- 30 db. This algorithm is an excellent choice for the spike classification.