NMR spectral quantitation by principal component analysis: III. A generalized procedure for determination of lineshape variations

R. Stoyanova, T. R. Brown

Research output: Contribution to journalArticle

27 Scopus citations

Abstract

We present a general procedure for automatic quantitation of a series of spectral peaks based on principal component analysis (PCA). PCA has been previously used for spectral quantitation of a single resonant peak of constant shape but variable amplitude. Here we extend this procedure to estimate all of the peak parameters: amplitude, position (frequency), phase and linewidth. The procedure consists of a series of iterative steps in which the estimates of position and phase from one stage of iteration are used to correct the spectra prior to the next stage. The process is convergent to a stable result, typically in less than 5 iterations. If desired, remaining linewidth variations can then be corrected. Correction of (typically) unwanted variations of these types is important not only for direct peak quantitation, but also as a preprocessing step for spectral data prior to application of pattern recognition/classification techniques. The procedure is demonstrated on simulated data and on a set of 992 31P NMR in vivo spectra taken from a kinetic study of rat muscle energetics. The proposed procedure is robust, makes very limited assumptions about the lineshape, and performs well with data of low signal-to-noise ratio.

Original languageEnglish (US)
Pages (from-to)163-175
Number of pages13
JournalJournal of Magnetic Resonance
Volume154
Issue number2
DOIs
StatePublished - Jan 1 2002
Externally publishedYes

Keywords

  • Automatic quantitation
  • Lineshape correction
  • Peak parameters
  • Principal component analysis
  • Spectroscopy

ASJC Scopus subject areas

  • Molecular Biology
  • Physical and Theoretical Chemistry
  • Spectroscopy
  • Radiology Nuclear Medicine and imaging
  • Condensed Matter Physics

Fingerprint Dive into the research topics of 'NMR spectral quantitation by principal component analysis: III. A generalized procedure for determination of lineshape variations'. Together they form a unique fingerprint.

  • Cite this