Spike Sorting Method Based on Two-Stage Radial Basis Function Networks - Advances in Computational Science and Engineering - page 98

Spike Sorting Method Based on Two-Stage Radial Basis
Function Networks

Jianhua Dai 1,2 , Xiaochun Liu 1,2 , Shaomin Zhang 1,3,4 , Huaijian Zhang 1,3,4 ,
Qingbo Wang 1,3,4 , and Xiaoxiang Zheng 1,3,4

1

Qiushi Academy for Advanced Studies, Zhejiang University, Hangzhou 310027, China
2 College of Computer Science and Technology, Zhejiang University,
Hangzhou 310027, China
3 College of Biomedical Engineering and Instrument Science,
Zhejiang University, Hangzhou 310027, China
4 Key Laboratory of Biomedical Engineering of Ministry of Education,
Zhejiang University, Hangzhou 310027, China
[e-mail]

Abstract. In this paper, 2-stage Radial Basis Function (RBF) Network method
is used for neural spike sorting. Firstly, raw signals are obtained from Neural
Signal Simulator, and added white noise ranged from -10dB to -40dB. Sec-
ondly, spikes are detected out with matched filter from signals. Lastly, 2-stage
RBF networks are constructed and the spikes are sorted using RBF networks.
The experiments show that 2-stage RBF network is an effective tool for neural
spike sorting.

Keywords: RBF, spike sorting, neural network, K-means.

1 Introduction

Classification of neural action potentials is a prerequisite to study neural activity and
brain functions. In extracellular recordings, a single electrode placed in the cortex
usually records action potentials from several neurons nearby. To obtain the firing
sequences of each neuron, spikes, which represent action potentials, has to be classified
by some methods. The goal of spike sorting is to find the firing positions in time from
each neuron. A lot of methods have been applied for spike sorting [1].
Radial basis function (RBF) networks are a type of artificial neural networks. It's
proved that RBF networks have powerful function approximation capabilities, good
local structure and efficient training algorithms. They have been successfully applied to
many fields such as speech pattern classification, image processing, nonlinear system
identification, nonlinear feature extraction etc. [2,3]. In this paper, we propose a
method for neural spike sorting task based on 2-stage RBF networks. The experiments
show that Radial Basis Function (RBF) Network is an effective tool for neural spike
sorting task.