Blind decomposition of infrared spectra using flexible component analysis

Abstract

The paper presents flexible component analysis-based blind decomposition of the mixtures of Fourier transform of infrared spectral (FT-IR) data into pure components, wherein the number of mixtures is less than number of pure components. The novelty of the proposed approach to blind FT-IR spectra decomposition is in use of hierarchical or local alternating least square nonnegative matrix factorization (HALS NMF) method with smoothness and sparseness constraints simultaneously imposed on the pure components. In contrast to many existing blind decomposition methods no a priori information about the number of pure components is required. It is estimated from the mixtures using robust data clustering algorithm in the wavelet domain. The HALS NMF method is compared favorably against 2 three sparse component analysis algorithms on experimental data with the known pure component spectra. Proposed methodology can be implemented as a part of software packages used for the analysis of FT-IR spectra and identification of chemical compounds.

Blind Separation of Analytes in Nuclear Magnetic Resonance Spectroscopy and Mass Spectrometry: Sparseness-Based Robust Multicomponent Analysis

“…Moreover, as stated by the authors, the BTEM approach is inapplicable when the number of experimentally measured spectra is less than the number of observed components. Here, as well as in recent publications, [16][17][18] we have demonstrated that the sparseness-based approach successfully estimates pure components when the number of available mixtures is less than the unknown number of components.…”

“…Thus, they are not applicable to the uBSS problem that is of central interest here. To estimate the number of analytes for an identified set of SAPs, we propose to use the clustering function: [16][17][18] ( ) ( )…”

Overconstrained library-based fitting method reveals age- and disease-related differences in transcutaneous Raman spectra of murine bones

“…Application of one or more a priori constraints (e.g., targeting a peak as in BTEM or overconstraining a fitting algorithm by using a fixed number of spectral components as in SOLD) can drive the estimation procedure. Blind decomposition methods, 55,56 if properly constrained, could also potentially be used to estimate bone spectra based on transcutaneous measurements. The results of this study demonstrate that the mineral/matrix ratio of cortical bone, which has frequently been used as a Raman-based indicator of bone health and strength, [4][5][6][7][8][9][10][11][12][13][14] can be measured transcutaneously by processing SORS data sets with SOLD ( Figs.…”

Component recognition with three-dimensional fluorescence spectra based on non-negative matrix factorization

“…These are exactly two features of the results obtained from NMF. As a result, NMF has been applied to the data analysis of multicomponent spectra . However, in most of the related literatures involving NMF, the analyzed spectra are two dimensional.…”