Fermentation process tracking through enhanced spectral calibration modeling

Sophia Triadaphillou, Elaine Martin, Gary Montague, Alison Nordon, Paul Jeffkins, Sarah Stimpson

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

The FDA process analytical technology (PAT) initiative will materialize in a significant increase in the number of installations of spectroscopic instrumentation: However, to attain the greatest benefit from the data generated, there is a need for calibration procedures that extract the maximum information content. For example, in fermentation processes, the interpretation of the resulting spectra is challenging as a consequence. of the large number of wavelengths recorded, the underlying correlation structure that is evident between the wavelengths and the impact of the measurement environment. Approaches to the development of calibration models have been based on the. application of partial least squares (PLS) either to the full spectral signature or to a subset of wavelengths. This paper presents a new approach to calibration modeling that combines a wavelength selection procedure, spectral window selection (SWS), where windows of wavelengths are automatically selected which are subsequently used as the basis of the calibration model. However, due to the non-uniqueness of the windows selected when the algorithm is executed repeatedly, multiple models are constructed and these are then combined using stacking thereby increasing the robustness of the final calibration model. The methodology is I applied to data generated during the monitoring of broth concentrations in an industrial fermentation process from on-line near-infrared (NIR) and mid-infrared (MIR) spectrometers. It is shown that the proposed calibration modeling procedure outperforms traditional calibration procedures, as well as enabling the identification of the critical regions of the spectra with regard to the fermentation process.

Original languageEnglish
Pages (from-to)554-567
Number of pages14
JournalBiotechnology and Bioengineering
Volume97
Issue number3
DOIs
Publication statusPublished - 15 Jun 2007

Fingerprint

Fermentation
Calibration
Wavelength
Infrared spectrometers
Least-Squares Analysis
Infrared radiation
Technology
Monitoring

Keywords

  • PLS
  • infra red spectroscopy
  • fermentation
  • monitoring
  • wavelength selection
  • calibration modeling
  • near-infrared spectroscopy
  • partial least-squares
  • regression
  • glucose
  • multivariate calibration

Cite this

Triadaphillou, S., Martin, E., Montague, G., Nordon, A., Jeffkins, P., & Stimpson, S. (2007). Fermentation process tracking through enhanced spectral calibration modeling. Biotechnology and Bioengineering, 97(3), 554-567. https://doi.org/10.1002/bit.21248
Triadaphillou, Sophia ; Martin, Elaine ; Montague, Gary ; Nordon, Alison ; Jeffkins, Paul ; Stimpson, Sarah. / Fermentation process tracking through enhanced spectral calibration modeling. In: Biotechnology and Bioengineering. 2007 ; Vol. 97, No. 3. pp. 554-567.
@article{c19d617311d141f38403c22412d44ed6,
title = "Fermentation process tracking through enhanced spectral calibration modeling",
abstract = "The FDA process analytical technology (PAT) initiative will materialize in a significant increase in the number of installations of spectroscopic instrumentation: However, to attain the greatest benefit from the data generated, there is a need for calibration procedures that extract the maximum information content. For example, in fermentation processes, the interpretation of the resulting spectra is challenging as a consequence. of the large number of wavelengths recorded, the underlying correlation structure that is evident between the wavelengths and the impact of the measurement environment. Approaches to the development of calibration models have been based on the. application of partial least squares (PLS) either to the full spectral signature or to a subset of wavelengths. This paper presents a new approach to calibration modeling that combines a wavelength selection procedure, spectral window selection (SWS), where windows of wavelengths are automatically selected which are subsequently used as the basis of the calibration model. However, due to the non-uniqueness of the windows selected when the algorithm is executed repeatedly, multiple models are constructed and these are then combined using stacking thereby increasing the robustness of the final calibration model. The methodology is I applied to data generated during the monitoring of broth concentrations in an industrial fermentation process from on-line near-infrared (NIR) and mid-infrared (MIR) spectrometers. It is shown that the proposed calibration modeling procedure outperforms traditional calibration procedures, as well as enabling the identification of the critical regions of the spectra with regard to the fermentation process.",
keywords = "PLS, infra red spectroscopy, fermentation, monitoring, wavelength selection, calibration modeling, near-infrared spectroscopy, partial least-squares, regression, glucose, multivariate calibration",
author = "Sophia Triadaphillou and Elaine Martin and Gary Montague and Alison Nordon and Paul Jeffkins and Sarah Stimpson",
year = "2007",
month = "6",
day = "15",
doi = "10.1002/bit.21248",
language = "English",
volume = "97",
pages = "554--567",
journal = "Biotechnology and Bioengineering",
issn = "0006-3592",
number = "3",

}

Triadaphillou, S, Martin, E, Montague, G, Nordon, A, Jeffkins, P & Stimpson, S 2007, 'Fermentation process tracking through enhanced spectral calibration modeling', Biotechnology and Bioengineering, vol. 97, no. 3, pp. 554-567. https://doi.org/10.1002/bit.21248

Fermentation process tracking through enhanced spectral calibration modeling. / Triadaphillou, Sophia; Martin, Elaine; Montague, Gary; Nordon, Alison; Jeffkins, Paul; Stimpson, Sarah.

In: Biotechnology and Bioengineering, Vol. 97, No. 3, 15.06.2007, p. 554-567.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fermentation process tracking through enhanced spectral calibration modeling

AU - Triadaphillou, Sophia

AU - Martin, Elaine

AU - Montague, Gary

AU - Nordon, Alison

AU - Jeffkins, Paul

AU - Stimpson, Sarah

PY - 2007/6/15

Y1 - 2007/6/15

N2 - The FDA process analytical technology (PAT) initiative will materialize in a significant increase in the number of installations of spectroscopic instrumentation: However, to attain the greatest benefit from the data generated, there is a need for calibration procedures that extract the maximum information content. For example, in fermentation processes, the interpretation of the resulting spectra is challenging as a consequence. of the large number of wavelengths recorded, the underlying correlation structure that is evident between the wavelengths and the impact of the measurement environment. Approaches to the development of calibration models have been based on the. application of partial least squares (PLS) either to the full spectral signature or to a subset of wavelengths. This paper presents a new approach to calibration modeling that combines a wavelength selection procedure, spectral window selection (SWS), where windows of wavelengths are automatically selected which are subsequently used as the basis of the calibration model. However, due to the non-uniqueness of the windows selected when the algorithm is executed repeatedly, multiple models are constructed and these are then combined using stacking thereby increasing the robustness of the final calibration model. The methodology is I applied to data generated during the monitoring of broth concentrations in an industrial fermentation process from on-line near-infrared (NIR) and mid-infrared (MIR) spectrometers. It is shown that the proposed calibration modeling procedure outperforms traditional calibration procedures, as well as enabling the identification of the critical regions of the spectra with regard to the fermentation process.

AB - The FDA process analytical technology (PAT) initiative will materialize in a significant increase in the number of installations of spectroscopic instrumentation: However, to attain the greatest benefit from the data generated, there is a need for calibration procedures that extract the maximum information content. For example, in fermentation processes, the interpretation of the resulting spectra is challenging as a consequence. of the large number of wavelengths recorded, the underlying correlation structure that is evident between the wavelengths and the impact of the measurement environment. Approaches to the development of calibration models have been based on the. application of partial least squares (PLS) either to the full spectral signature or to a subset of wavelengths. This paper presents a new approach to calibration modeling that combines a wavelength selection procedure, spectral window selection (SWS), where windows of wavelengths are automatically selected which are subsequently used as the basis of the calibration model. However, due to the non-uniqueness of the windows selected when the algorithm is executed repeatedly, multiple models are constructed and these are then combined using stacking thereby increasing the robustness of the final calibration model. The methodology is I applied to data generated during the monitoring of broth concentrations in an industrial fermentation process from on-line near-infrared (NIR) and mid-infrared (MIR) spectrometers. It is shown that the proposed calibration modeling procedure outperforms traditional calibration procedures, as well as enabling the identification of the critical regions of the spectra with regard to the fermentation process.

KW - PLS

KW - infra red spectroscopy

KW - fermentation

KW - monitoring

KW - wavelength selection

KW - calibration modeling

KW - near-infrared spectroscopy

KW - partial least-squares

KW - regression

KW - glucose

KW - multivariate calibration

U2 - 10.1002/bit.21248

DO - 10.1002/bit.21248

M3 - Article

VL - 97

SP - 554

EP - 567

JO - Biotechnology and Bioengineering

JF - Biotechnology and Bioengineering

SN - 0006-3592

IS - 3

ER -