Quantification and handling of nonlinearity in Raman micro-spectrometry of pharmaceuticals

Brigitta Nagy, Attila Farkas, Attila Balogh, Hajnalka Pataki, Balázs Vajna, Zsombor K. Nagy, G. Marosi

Research output: Article

4 Citations (Scopus)

Abstract

This work demonstrates how nonlinearity in Raman spectrometry of pharmaceuticals can be handled and accurate quantification can be achieved by applying certain chemometric methods including variable selection. Such approach proved to be successful even if the component spectra are very similar or spectral intensities of the constituents are strongly different. The relevant examples are: blends of two crystalline forms of carvedilol ("CRYST-PM" blend) and a three-component pharmaceutical model system ("PHARM-TM" blend). The widely used classical least squares regression (CLS) and partial least squares regression (PLS) quantification methods provided relatively poor root mean squared error of prediction (RMSEP) values: approximately 2-4% and 4-10% for CRYST-PM and PHARM-TM respectively. The residual plots of these models indicated the nonlinearity of the preprocessed data sets. More accurate quantitative results could be achieved with properly applied variable selection methods. It was observed that variable selection methods discarded the most intensive bands while less intensive ones were retained as the most informative spectral ranges. As a result not only the accuracy of concentration determination was enhanced, but the linearity of models was improved as well. This indicated that nonlinearity occurred especially at the intensive spectral bands. Other methods developed for handling nonlinearity were also capable of adapting to the spectral nature of both data sets. The RMSEP could be decreased this way to 1% in CRYST-PM and 3-6% in PHARM-TM. Raman maps with accurate real concentrations could be prepared this way. All quantitative models were compared by the non-parametric sum of ranking differences (SRD) method, which also proved that models based on variable selection or nonlinear methods provide better quantification.

Original languageEnglish
Pages (from-to)236-246
Number of pages11
JournalJournal of Pharmaceutical and Biomedical Analysis
Volume128
DOIs
Publication statusPublished - szept. 5 2016

Fingerprint

Spectrometry
Spectrum Analysis
Pharmaceutical Preparations
Least-Squares Analysis
Crystalline materials

ASJC Scopus subject areas

  • Analytical Chemistry
  • Drug Discovery
  • Pharmaceutical Science
  • Spectroscopy
  • Clinical Biochemistry

Cite this

Quantification and handling of nonlinearity in Raman micro-spectrometry of pharmaceuticals. / Nagy, Brigitta; Farkas, Attila; Balogh, Attila; Pataki, Hajnalka; Vajna, Balázs; Nagy, Zsombor K.; Marosi, G.

In: Journal of Pharmaceutical and Biomedical Analysis, Vol. 128, 05.09.2016, p. 236-246.

Research output: Article

Nagy, Brigitta ; Farkas, Attila ; Balogh, Attila ; Pataki, Hajnalka ; Vajna, Balázs ; Nagy, Zsombor K. ; Marosi, G. / Quantification and handling of nonlinearity in Raman micro-spectrometry of pharmaceuticals. In: Journal of Pharmaceutical and Biomedical Analysis. 2016 ; Vol. 128. pp. 236-246.
@article{6f1bfbf1e00c41b5ab3c32d4563ee59e,
title = "Quantification and handling of nonlinearity in Raman micro-spectrometry of pharmaceuticals",
abstract = "This work demonstrates how nonlinearity in Raman spectrometry of pharmaceuticals can be handled and accurate quantification can be achieved by applying certain chemometric methods including variable selection. Such approach proved to be successful even if the component spectra are very similar or spectral intensities of the constituents are strongly different. The relevant examples are: blends of two crystalline forms of carvedilol ({"}CRYST-PM{"} blend) and a three-component pharmaceutical model system ({"}PHARM-TM{"} blend). The widely used classical least squares regression (CLS) and partial least squares regression (PLS) quantification methods provided relatively poor root mean squared error of prediction (RMSEP) values: approximately 2-4{\%} and 4-10{\%} for CRYST-PM and PHARM-TM respectively. The residual plots of these models indicated the nonlinearity of the preprocessed data sets. More accurate quantitative results could be achieved with properly applied variable selection methods. It was observed that variable selection methods discarded the most intensive bands while less intensive ones were retained as the most informative spectral ranges. As a result not only the accuracy of concentration determination was enhanced, but the linearity of models was improved as well. This indicated that nonlinearity occurred especially at the intensive spectral bands. Other methods developed for handling nonlinearity were also capable of adapting to the spectral nature of both data sets. The RMSEP could be decreased this way to 1{\%} in CRYST-PM and 3-6{\%} in PHARM-TM. Raman maps with accurate real concentrations could be prepared this way. All quantitative models were compared by the non-parametric sum of ranking differences (SRD) method, which also proved that models based on variable selection or nonlinear methods provide better quantification.",
keywords = "Chemometrics, Hyperspectral imaging, Multivariate regression, Nonlinear behavior, Pharmaceutical analysis, Variable selection",
author = "Brigitta Nagy and Attila Farkas and Attila Balogh and Hajnalka Pataki and Bal{\'a}zs Vajna and Nagy, {Zsombor K.} and G. Marosi",
year = "2016",
month = "9",
day = "5",
doi = "10.1016/j.jpba.2016.05.036",
language = "English",
volume = "128",
pages = "236--246",
journal = "Journal of Pharmaceutical and Biomedical Analysis",
issn = "0731-7085",
publisher = "Elsevier",

}

TY - JOUR

T1 - Quantification and handling of nonlinearity in Raman micro-spectrometry of pharmaceuticals

AU - Nagy, Brigitta

AU - Farkas, Attila

AU - Balogh, Attila

AU - Pataki, Hajnalka

AU - Vajna, Balázs

AU - Nagy, Zsombor K.

AU - Marosi, G.

PY - 2016/9/5

Y1 - 2016/9/5

N2 - This work demonstrates how nonlinearity in Raman spectrometry of pharmaceuticals can be handled and accurate quantification can be achieved by applying certain chemometric methods including variable selection. Such approach proved to be successful even if the component spectra are very similar or spectral intensities of the constituents are strongly different. The relevant examples are: blends of two crystalline forms of carvedilol ("CRYST-PM" blend) and a three-component pharmaceutical model system ("PHARM-TM" blend). The widely used classical least squares regression (CLS) and partial least squares regression (PLS) quantification methods provided relatively poor root mean squared error of prediction (RMSEP) values: approximately 2-4% and 4-10% for CRYST-PM and PHARM-TM respectively. The residual plots of these models indicated the nonlinearity of the preprocessed data sets. More accurate quantitative results could be achieved with properly applied variable selection methods. It was observed that variable selection methods discarded the most intensive bands while less intensive ones were retained as the most informative spectral ranges. As a result not only the accuracy of concentration determination was enhanced, but the linearity of models was improved as well. This indicated that nonlinearity occurred especially at the intensive spectral bands. Other methods developed for handling nonlinearity were also capable of adapting to the spectral nature of both data sets. The RMSEP could be decreased this way to 1% in CRYST-PM and 3-6% in PHARM-TM. Raman maps with accurate real concentrations could be prepared this way. All quantitative models were compared by the non-parametric sum of ranking differences (SRD) method, which also proved that models based on variable selection or nonlinear methods provide better quantification.

AB - This work demonstrates how nonlinearity in Raman spectrometry of pharmaceuticals can be handled and accurate quantification can be achieved by applying certain chemometric methods including variable selection. Such approach proved to be successful even if the component spectra are very similar or spectral intensities of the constituents are strongly different. The relevant examples are: blends of two crystalline forms of carvedilol ("CRYST-PM" blend) and a three-component pharmaceutical model system ("PHARM-TM" blend). The widely used classical least squares regression (CLS) and partial least squares regression (PLS) quantification methods provided relatively poor root mean squared error of prediction (RMSEP) values: approximately 2-4% and 4-10% for CRYST-PM and PHARM-TM respectively. The residual plots of these models indicated the nonlinearity of the preprocessed data sets. More accurate quantitative results could be achieved with properly applied variable selection methods. It was observed that variable selection methods discarded the most intensive bands while less intensive ones were retained as the most informative spectral ranges. As a result not only the accuracy of concentration determination was enhanced, but the linearity of models was improved as well. This indicated that nonlinearity occurred especially at the intensive spectral bands. Other methods developed for handling nonlinearity were also capable of adapting to the spectral nature of both data sets. The RMSEP could be decreased this way to 1% in CRYST-PM and 3-6% in PHARM-TM. Raman maps with accurate real concentrations could be prepared this way. All quantitative models were compared by the non-parametric sum of ranking differences (SRD) method, which also proved that models based on variable selection or nonlinear methods provide better quantification.

KW - Chemometrics

KW - Hyperspectral imaging

KW - Multivariate regression

KW - Nonlinear behavior

KW - Pharmaceutical analysis

KW - Variable selection

UR - http://www.scopus.com/inward/record.url?scp=84973146903&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84973146903&partnerID=8YFLogxK

U2 - 10.1016/j.jpba.2016.05.036

DO - 10.1016/j.jpba.2016.05.036

M3 - Article

C2 - 27281579

AN - SCOPUS:84973146903

VL - 128

SP - 236

EP - 246

JO - Journal of Pharmaceutical and Biomedical Analysis

JF - Journal of Pharmaceutical and Biomedical Analysis

SN - 0731-7085

ER -