Title: Eigenspectra, a robust regression method for multiplexed Raman spectra analysis

Authors: Shuo Li; James O. Nyagilo; Digant P. Dave; Baoju Zhang; Jean Gao

Addresses: Computer Science and Engineering Department, University of Texas at Arlington, Arlington, Texas 76013, USA ' Bioengineering Department, University of Texas at Arlington, Arlington, Texas 76013, USA ' Bioengineering Department, University of Texas at Arlington, Arlington, Texas 76013, USA ' College of Physics and Information Science, Tianjin Normal University, Tianjin 300387, China ' Computer Science and Engineering Department, University of Texas at Arlington, Arlington, Texas 76013, USA

Abstract: With the latest development of Surface Enhanced Raman Scattering (SERS) nanoparticles, Raman spectroscopy now can be extended to bioimaging and biosensing. In this study, we demonstrate the ability of Raman spectroscopy to separate multiple spectral fingerprints using Raman nanotags. A machine learning method is proposed to estimate the mixing ratios of sources from mixture signals. It decomposes the mixture signals into components for both best representation and most relating to mixing ratios. Then regression coefficients are calculated for the prediction. The robustness of the method was compared with least squares and weighted least squares methods.

Keywords: quantitative analysis; Raman spectroscopy; component decomposition; eigenspectra; data mining; bioinformatics; robust regression; multiplexed Raman spectra analysis; bioimaging; biosensing; SERS nanoparticles; Raman nanotags; multiple spectral fingerprints; machine learning; nanotechnology.

DOI: 10.1504/IJDMB.2013.054224

International Journal of Data Mining and Bioinformatics, 2013 Vol.7 No.4, pp.358 - 375

Received: 11 May 2011
Accepted: 11 May 2011
Published online: 20 Oct 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article