Comprehensive assessment of prediction of pH in red meat based on near-infrared spectroscopy (NIRS)
Abstract
pH is an indicator of meat quality, affecting colour, tenderness, and shelf life. This study is aimed to carry a comprehensive assessment of non-invasive prediction model for pH using Near-Infrared Spectroscopy (NIRS) across red meat species. A calibration model based on 2539 samples was built using a deep convolutional neural network (DCNN), integrating data from 17 muscles, 3 species (beef, lamb, venison), and 4 NIRS devices. The model achieved a strong calibration performance (R 2 = 0.95, SEP = 0.07) and a competent validation performance (R 2 = 0.82, SEP = 0.14). External testing confirmed the model's ability to predict pH across different animals and muscles, although reduced accuracy was observed with the INDIGO handheld device. The lack of ability to capture information related to muscle structure reduced the prediction performance of the sensor and generated bias. Results confirm the feasibility of a robust NIRS-based tool for meat pH assessment, provided device used to collect NIR spectra is suitable for red meat. This approach opens new opportunities for real-time quality control in meat processing, with a potential to improve consistency and reduce losses due to pH-related defects.
Related
- Near-infrared spectroscopy for rapid and non-destructive assessment of PFAS effects in blue mussels (Mytilus spp.): a proof-of-concept study
- Monitoring a Polyurethane Synthesis by Fiber-Coupled Attenuated Total Reflection Fourier Transform Infrared Spectroscopy and Multivariate Curve Resolution-Alternating Least Squares
- Solvent-Dependent Self-Association of Chiral 2-Acetoxy-2-Phenylacetic Acid Revealed by Infrared and Vibrational Circular Dichroism Spectroscopy
- Gas-Phase Far-Infrared and Rotational Spectroscopy of 1-and 2-Cyanonaphthalene: Experiment and Theory
- Rapid Determination of Soybean Protein Content by Near-Infrared Spectroscopy Coupled with Multi-Learner Ensemble Wavelength Selection
BibTeX
@article{elliesoury2026comprehensive,
title = {{Comprehensive assessment of prediction of pH in red meat based on near-infrared spectroscopy (NIRS)}},
author = {Marie-Pierre Ellies-Oury and C.E. Realini and R. Zhang and Y. Dixit and M.M. Reis},
journal = {Meat Science},
year = {2026},
month = oct,
doi = {10.1016/j.meatsci.2026.110240},
url = {https://hal.inrae.fr/hal-05839578v1},
note = {CC BY},
}
RIS
TY - JOUR
TI - Comprehensive assessment of prediction of pH in red meat based on near-infrared spectroscopy (NIRS)
AU - Marie-Pierre Ellies-Oury
AU - C.E. Realini
AU - R. Zhang
AU - Y. Dixit
AU - M.M. Reis
PY - 2026
DA - 2026/10/26
JO - Meat Science
DO - 10.1016/j.meatsci.2026.110240
UR - https://hal.inrae.fr/hal-05839578v1
AB - pH is an indicator of meat quality, affecting colour, tenderness, and shelf life. This study is aimed to carry a comprehensive assessment of non-invasive prediction model for pH using Near-Infrared Spectroscopy (NIRS) across red meat species. A calibration model based on 2539 samples was built using a deep convolutional neural network (DCNN), integrating data from 17 muscles, 3 species (beef, lamb, venison), and 4 NIRS devices. The model achieved a strong calibration performance (R 2 = 0.95, SEP = 0.07) and a competent validation performance (R 2 = 0.82, SEP = 0.14). External testing confirmed the model's ability to predict pH across different animals and muscles, although reduced accuracy was observed with the INDIGO handheld device. The lack of ability to capture information related to muscle structure reduced the prediction performance of the sensor and generated bias. Results confirm the feasibility of a robust NIRS-based tool for meat pH assessment, provided device used to collect NIR spectra is suitable for red meat. This approach opens new opportunities for real-time quality control in meat processing, with a potential to improve consistency and reduce losses due to pH-related defects.
SN - 0309-1740
SN - 0309-1740
ER -