压缩感知的重叠神经元峰电位信号分类

Compressive Sensing for Overlapping Neural Spike Sorting

随着电生理技术水平的提高,电极可以记录的峰电位信号包含多个神经元峰电位的叠加.本文提出了一种采用压缩感知和最大后验估计的分类算法来解决重叠峰电位分类问题.其中压缩感知算法用于得到稀疏信号,最大后验估计用于搜索出稀疏信号的最优解.在实验中,我们采用仿真和实测的三组数据对本文算法和传统算法进行了测试,实验结果表明,当峰电位波形相似时,相比于k-均值聚类及CBP(Continuous Basis Pursuit)算法,本文算法具有较少的分类错误数.

With the development of electrophysiological technology,the spike signals that electrodes record contain multi-neuron overlapped spikes.This paper presents a classification method based on a compressed sensing algorithm and a maximum a posteriori(MAP)estimate to sort the overlapped spikes.The compressed sensing algorithm is used to obtain sparse signals,and the maximum a posteriori estimate is used to search an optimal value in the sparse signals.In experiments,we use one group of simulation data and two groups of measured data to verify the method.The experimental results show that when the spike waveform shapes in the data are similar,the proposed method has fewer sorting errors compared with the existing algorithms,k-means clustering and CBP(Continuous Basis Pursuit).