Maintaining the predictive abilities of multivariate calibration models by spectral space transformation

Wen Du, Zeng-Ping Chen, Li-Jing Zhong, Shu-Xia Wang, Ru-Qin Yu, Alison Nordon, David Littlejohn, Megan Holden

Research output: Contribution to journalArticle

49 Citations (Scopus)

Abstract

In quantitative on-line/in-line monitoring of chemical and bio-chemical processes using spectroscopic instruments, multivariate calibration models are indispensable for the extraction of chemical information from complex spectroscopic measurements. The development of reliable multivariate calibration models is generally time-consuming and costly. Therefore, once a reliable multivariate calibration model is established, it is expected to be used for an extended period. However, any change in the instrumental response or variations in the measurement conditions can renders multivariate calibration model invalid. In this contribution, a new method, spectral space transformation (SST), has been developed to maintain the predictive abilities of multivariate calibration models when the spectrometer or measurement conditions are altered. SST tries to eliminate the spectral differences induced by the changes in instruments or measurement conditions through the transformation between two spectral spaces spanned by the corresponding spectra of a subset of standardization samples measured on two instruments or under two sets of experimental conditions. The performance of the method has been tested on two data sets comprising NIR and MIR spectra. The experimental results show that SST can achieve satisfactory analyte predictions from spectroscopic measurements subject to spectrometer/probe alteration, when only a few standardization samples are used. Compared with the existing popular methods designed for the same purpose. i.e. global PLS, univariate slope and bias correction (SBC) and piecewise direct standardization (PDS). SST has the advantages of implementation simplicity, wider applicability and better performance in terms of predictive accuracy.

LanguageEnglish
Pages64-70
Number of pages7
JournalAnalytica Chimica Acta
Volume690
Issue number1
Early online date12 Feb 2011
DOIs
Publication statusPublished - 25 Mar 2011

Fingerprint

Space Simulation
Calibration
calibration
standardization
Standardization
Spectrometers
spectrometer
Chemical Phenomena
Environmental Monitoring
Information Storage and Retrieval
probe
Monitoring
monitoring
prediction
method

Keywords

  • calibration model maintenance
  • spectral space transformation
  • spectral standardization
  • spectroscopic instruments
  • process analytical technology
  • infrared spectrometric instruments
  • standardization procedure
  • genetic regression
  • raman-spectroscopy
  • neural network
  • improvement
  • algorithm
  • transferability
  • linearity

Cite this

Du, Wen ; Chen, Zeng-Ping ; Zhong, Li-Jing ; Wang, Shu-Xia ; Yu, Ru-Qin ; Nordon, Alison ; Littlejohn, David ; Holden, Megan. / Maintaining the predictive abilities of multivariate calibration models by spectral space transformation. In: Analytica Chimica Acta. 2011 ; Vol. 690, No. 1. pp. 64-70.
@article{05175ba283aa4bb2851337aaacb40c16,
title = "Maintaining the predictive abilities of multivariate calibration models by spectral space transformation",
abstract = "In quantitative on-line/in-line monitoring of chemical and bio-chemical processes using spectroscopic instruments, multivariate calibration models are indispensable for the extraction of chemical information from complex spectroscopic measurements. The development of reliable multivariate calibration models is generally time-consuming and costly. Therefore, once a reliable multivariate calibration model is established, it is expected to be used for an extended period. However, any change in the instrumental response or variations in the measurement conditions can renders multivariate calibration model invalid. In this contribution, a new method, spectral space transformation (SST), has been developed to maintain the predictive abilities of multivariate calibration models when the spectrometer or measurement conditions are altered. SST tries to eliminate the spectral differences induced by the changes in instruments or measurement conditions through the transformation between two spectral spaces spanned by the corresponding spectra of a subset of standardization samples measured on two instruments or under two sets of experimental conditions. The performance of the method has been tested on two data sets comprising NIR and MIR spectra. The experimental results show that SST can achieve satisfactory analyte predictions from spectroscopic measurements subject to spectrometer/probe alteration, when only a few standardization samples are used. Compared with the existing popular methods designed for the same purpose. i.e. global PLS, univariate slope and bias correction (SBC) and piecewise direct standardization (PDS). SST has the advantages of implementation simplicity, wider applicability and better performance in terms of predictive accuracy.",
keywords = "calibration model maintenance, spectral space transformation, spectral standardization, spectroscopic instruments, process analytical technology, infrared spectrometric instruments, standardization procedure, genetic regression, raman-spectroscopy, neural network, improvement, algorithm, transferability, linearity",
author = "Wen Du and Zeng-Ping Chen and Li-Jing Zhong and Shu-Xia Wang and Ru-Qin Yu and Alison Nordon and David Littlejohn and Megan Holden",
year = "2011",
month = "3",
day = "25",
doi = "10.1016/j.aca.2011.02.014",
language = "English",
volume = "690",
pages = "64--70",
journal = "Analytica Chimica Acta",
issn = "0003-2670",
number = "1",

}

Maintaining the predictive abilities of multivariate calibration models by spectral space transformation. / Du, Wen; Chen, Zeng-Ping; Zhong, Li-Jing; Wang, Shu-Xia; Yu, Ru-Qin; Nordon, Alison; Littlejohn, David; Holden, Megan.

In: Analytica Chimica Acta, Vol. 690, No. 1, 25.03.2011, p. 64-70.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Maintaining the predictive abilities of multivariate calibration models by spectral space transformation

AU - Du, Wen

AU - Chen, Zeng-Ping

AU - Zhong, Li-Jing

AU - Wang, Shu-Xia

AU - Yu, Ru-Qin

AU - Nordon, Alison

AU - Littlejohn, David

AU - Holden, Megan

PY - 2011/3/25

Y1 - 2011/3/25

N2 - In quantitative on-line/in-line monitoring of chemical and bio-chemical processes using spectroscopic instruments, multivariate calibration models are indispensable for the extraction of chemical information from complex spectroscopic measurements. The development of reliable multivariate calibration models is generally time-consuming and costly. Therefore, once a reliable multivariate calibration model is established, it is expected to be used for an extended period. However, any change in the instrumental response or variations in the measurement conditions can renders multivariate calibration model invalid. In this contribution, a new method, spectral space transformation (SST), has been developed to maintain the predictive abilities of multivariate calibration models when the spectrometer or measurement conditions are altered. SST tries to eliminate the spectral differences induced by the changes in instruments or measurement conditions through the transformation between two spectral spaces spanned by the corresponding spectra of a subset of standardization samples measured on two instruments or under two sets of experimental conditions. The performance of the method has been tested on two data sets comprising NIR and MIR spectra. The experimental results show that SST can achieve satisfactory analyte predictions from spectroscopic measurements subject to spectrometer/probe alteration, when only a few standardization samples are used. Compared with the existing popular methods designed for the same purpose. i.e. global PLS, univariate slope and bias correction (SBC) and piecewise direct standardization (PDS). SST has the advantages of implementation simplicity, wider applicability and better performance in terms of predictive accuracy.

AB - In quantitative on-line/in-line monitoring of chemical and bio-chemical processes using spectroscopic instruments, multivariate calibration models are indispensable for the extraction of chemical information from complex spectroscopic measurements. The development of reliable multivariate calibration models is generally time-consuming and costly. Therefore, once a reliable multivariate calibration model is established, it is expected to be used for an extended period. However, any change in the instrumental response or variations in the measurement conditions can renders multivariate calibration model invalid. In this contribution, a new method, spectral space transformation (SST), has been developed to maintain the predictive abilities of multivariate calibration models when the spectrometer or measurement conditions are altered. SST tries to eliminate the spectral differences induced by the changes in instruments or measurement conditions through the transformation between two spectral spaces spanned by the corresponding spectra of a subset of standardization samples measured on two instruments or under two sets of experimental conditions. The performance of the method has been tested on two data sets comprising NIR and MIR spectra. The experimental results show that SST can achieve satisfactory analyte predictions from spectroscopic measurements subject to spectrometer/probe alteration, when only a few standardization samples are used. Compared with the existing popular methods designed for the same purpose. i.e. global PLS, univariate slope and bias correction (SBC) and piecewise direct standardization (PDS). SST has the advantages of implementation simplicity, wider applicability and better performance in terms of predictive accuracy.

KW - calibration model maintenance

KW - spectral space transformation

KW - spectral standardization

KW - spectroscopic instruments

KW - process analytical technology

KW - infrared spectrometric instruments

KW - standardization procedure

KW - genetic regression

KW - raman-spectroscopy

KW - neural network

KW - improvement

KW - algorithm

KW - transferability

KW - linearity

U2 - 10.1016/j.aca.2011.02.014

DO - 10.1016/j.aca.2011.02.014

M3 - Article

VL - 690

SP - 64

EP - 70

JO - Analytica Chimica Acta

T2 - Analytica Chimica Acta

JF - Analytica Chimica Acta

SN - 0003-2670

IS - 1

ER -