Efficient online feature extraction algorithm for spike sorting in a multichannel FPGA-based neural recording system

Author

Peng Li ; Ming Liu ; Xu Zhang ; Hongda Chen

Author_Institution

State Key Lab. on Integrated Optoelectron., Inst. of Semicond., Beijing, China

fYear

2014

fDate

22-24 Oct. 2014

Firstpage

1

Lastpage

4

Abstract

A novel feature extraction algorithm for multichannel FPGA-based neural recording systems is presented in this paper. It contains the Dual Vertex Threshold (DVT) and the Minimum Delimitation (MD), which are used for spike detection and feature vector extraction respectively. By reducing the computational complexity of DVT and MD, the difficulty of this algorithm in application is greatly reduced. Based on this characteristic, a multichannel FPGA hardware architecture is implemented in this paper. Using extracted feature vectors, the sorting performance of K-means is as good as that with the PCA-based features. Additionally, the test result shows that the transmission bandwidth is reduced to 1.62% of original data rate.

Keywords

computational complexity; feature extraction; field programmable gate arrays; medical signal detection; microelectrodes; neurophysiology; principal component analysis; DVT; Dual Vertex Threshold; K-means; MD; Minimum Delimitation; PCA-based feature; computational complexity; feature vector extraction; multichannel FPGA hardware architecture; multichannel FPGA-based neural recording systems; online feature extraction algorithm; original data rate; sorting performance; spike detection; spike sorting; transmission bandwidth; Feature extraction; Multiplexing; Principal component analysis; Sorting; Support vector machine classification; FPGA; Feature extraction; high efficiency; low complexity;

fLanguage

English

Publisher

ieee

Conference_Titel

Biomedical Circuits and Systems Conference (BioCAS), 2014 IEEE

Conference_Location

Lausanne

Type

conf

DOI

10.1109/BioCAS.2014.6981630

Filename

6981630