中文摘要：

爆发式锋电位（Burst）是大脑神经元动作电位发放的一种常见形式,它在增强神经信号传递的可靠性以及形成突触可塑性变化等方面具有重要的作用。在细胞外记录的锋电位信号中,同源Burst也表现为幅值和波形都明显变化的一串高频发放序列,这给神经元序列的正确分类提出了难题。为了解决这个问题,本研究设计了一种四极电极记录的锋电位信号检测和分类方法。在阈值法检出锋电位的基础上,首先根据锋电位时间间隔指标检出候选Burst信号小段,然后利用独立分量分析（ICA）的盲源分离特性,区分每个小段信号中所包含的不同来源的锋电位,再用于最后的整体信号的锋电位聚类。实验记录数据和仿真数据的检验结果表明,该方法不仅能够将来自不同神经元的Burst和单发放锋电位正确分类;而且,由于ICA应用的对象是短时间的候选Burst信号,因此,即使对于4通道信号也能够满足ICA源信号数量小于记录信号通道数的限制条件,同时,短信号处理又避免了ICA计算量大等问题,为Burst的正确检测与分类提供了一种新方法。

英文摘要：

Complex spike burst（i.e.,Burst） is one of the common modes of neuronal action potential firing in brain.Evidences have showed that burst firings play important roles in increasing reliability of neuronal signal transmission,in generating synaptic plasticity and so on.In extracellular recordings,a burst appears as a train of high-frequency firing spikes with obvious changes both in amplitude and in waveform of spikes.The non-steady change feature of burst spikes is a challenge to the accurate analysis of neuronal firing sequences.This paper presented a spike sorting algorithm for tetrode recording signals in order to deal with the burst spikes.In this method,based on the spike signals collected by a threshold detecting method,short candidate burst segments were selected firstly according to inter-spike intervals.Then the independent component analysis（ICA） was used to separate spikes from different sources in each of the short signal segments.Finally spike clustering for the whole signals was fulfilled based on the results of ICA.The results obtained from both experimental recordings and synthetic data showed that the algorithm was able to sort the burst spikes and single spikes from different sources accurately.In addition,because the ICA is applied on the very short signal segments of candidate bursts,even for the only four channel signals the algorithm can also meet the source signal number limit of ICA technique.Further more,the short signal processing decreases the ICA computing time.Therefore,the algorithm provides a new method for accurate burst spike detecting and sorting.