Evaluating low- mid- and high-level fusion strategies for combining Raman and infrared spectroscopy for quality assessment of red meat

Chima Robert; William Jessep; Joshua J Sutton; Talia M Hicks; Mark Loeffen; Mustafa Farouk; James F Ward; Wendy E Bain; Cameron R Craigie; Sara J Fraser-Miller; Keith C Gordon

doi:10.1016/j.foodchem.2021.130154

Evaluating low- mid- and high-level fusion strategies for combining Raman and infrared spectroscopy for quality assessment of red meat

Food Chem. 2021 Nov 1:361:130154. doi: 10.1016/j.foodchem.2021.130154. Epub 2021 May 18.

Authors

Affiliations

¹ Dodd-Walls Centre for Photonics and Quantum Technologies, Department of Chemistry, University of Otago, PO Box 56, Dunedin 9054, New Zealand.
² AgResearch, Grasslands Research Centre, Private Bag 11008, Palmerston North 4410, New Zealand.
³ Delytics Ltd., Waikato Innovation Centre, Hamilton East, Hamilton 3216, New Zealand.
⁴ AgResearch, Ruakura Research Centre, Private Bag 3123, Hamilton 3240, New Zealand.
⁵ AgResearch, Invermay Research Centre, Private Bag 50034, Mosgiel 9053, New Zealand.
⁶ AgResearch, Lincoln Research Centre, Private Bag 4749, Christchurch 8140, New Zealand.
⁷ Dodd-Walls Centre for Photonics and Quantum Technologies, Department of Chemistry, University of Otago, PO Box 56, Dunedin 9054, New Zealand. Electronic address: sara.miller@otago.ac.nz.
⁸ Dodd-Walls Centre for Photonics and Quantum Technologies, Department of Chemistry, University of Otago, PO Box 56, Dunedin 9054, New Zealand. Electronic address: keith.gordon@otago.ac.nz.

PMID: 34077882
DOI: 10.1016/j.foodchem.2021.130154

Abstract

The implementation of Raman and infrared spectroscopy with three data fusion strategies to predict pH and % IMF content of red meat was investigated. Raman and FTIR systems were utilized to assess quality parameters of intact red meat. Quantitative models were built using PLS, with model performances assessed with respect to the determination coefficient (R²), root mean square error and normalized root mean square error (NRMSEP). Results obtained on validation against an independent test set show that the high-level fusion strategy had the best performance in predicting the observed pH; with R_P² and NRMSEP values of 0.73 and 12.9% respectively, whereas low-level fusion strategy showed promise in predicting % IMF (NRMSEP = 8.5%). The fusion of data from more than one technique at low and high level resulted in improvement in the model performances; highlighting the possibility of information enhancement.

Keywords: % IMF; Chemometrics; Data fusion; Infrared spectroscopy; Raman spectroscopy; Red meat; pH.

MeSH terms

Animals
Food Analysis / methods*
Food Quality
Hydrogen-Ion Concentration
Red Meat / analysis*
Signal Processing, Computer-Assisted
Spectroscopy, Fourier Transform Infrared / methods*
Spectrum Analysis, Raman / methods*