- You are here:
-
-
All publications
- Real-time Prediction of Quality of Plant-based Meat Analogues Using Portable Hyperspectral Imaging System and Multivariate Analysis
All publications from your selection
Real-time Prediction of Quality of Plant-based Meat Analogues Using Portable Hyperspectral Imaging System and Multivariate Analysis
Authors: Logesh Dhanapal,
Chyngyz Erkinbaev
Identifier: CSBE23178
Download file:
Published in: CSBE-SCGAB Technical Conferences » AGM Lethbridge 2023
Download RAW file:
DOI:
Description: Plant-based meat analogues (PBMA) have received widescale research interest, and several reports have documented novel structuring techniques and techno-functional ingredients. However, there is a dearth of research concerning their quality and safety. Presence of numerous plant-derived ingredients with diverse physicochemical properties causes quality changes during storage which hinders their palatability and commercialization. Most existing wet chemistry approaches for quality monitoring are laborious, destructive, and time-consuming. This study reports the use of a non-destructive optical method using visible-near-infrared (VNIR) (400-1000 nm) portable hyperspectral imaging (HSI) coupled with multivariate analysis to monitor and predict the major quality traits of plant-based meat burgers (PBMB) including color, moisture, pH, and textural properties. HSI and quality measurements of 10 different PBMB formulations prepared with ingredient concentrations (w/w) ranging from 10-30% textured vegetable protein and 5-25% pea protein were recorded during a 14-day storage period. Spectral profiles extracted from VNIR-HSI were pre-processed with Savitzky?Golay 1st derivative with 11 smoothing points and mean centering. Unsupervised Principal Component Analysis (PCA) model on pre-processed data classified 140 PBMB samples based on storage days, formulations, and explored 10 wavelengths significantly contributing to predict their quality. Partial Least Square Regression (PLSR) was conducted both in the pre-processed full spectral range and selected wavelengths, to predict the quality traits. The figures of merit of PLSR models yielded good prediction of pH, redness, moisture, and hardness with low error values. The results unravel the feasibility of VNIR-HSI as a real-time and non-invasive method for predicting potential chemical changes of PBMA during storage.
Keywords: Plant-based meat, Hyperspectral imaging, Food quality, Plant proteins, Multivariate analysis
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2023-07-23
Technical field:
Conference name: CSBE/SCGAB 2023 Annual Conference, Lethbridge, Alberta, 23-26 July 2023.
Session name:
Other information:
Type: Presentation
Format:
Publication type: Text.Abstract
Source:
Relation:
Coverage: North_America
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
DOI:
Description: Plant-based meat analogues (PBMA) have received widescale research interest, and several reports have documented novel structuring techniques and techno-functional ingredients. However, there is a dearth of research concerning their quality and safety. Presence of numerous plant-derived ingredients with diverse physicochemical properties causes quality changes during storage which hinders their palatability and commercialization. Most existing wet chemistry approaches for quality monitoring are laborious, destructive, and time-consuming. This study reports the use of a non-destructive optical method using visible-near-infrared (VNIR) (400-1000 nm) portable hyperspectral imaging (HSI) coupled with multivariate analysis to monitor and predict the major quality traits of plant-based meat burgers (PBMB) including color, moisture, pH, and textural properties. HSI and quality measurements of 10 different PBMB formulations prepared with ingredient concentrations (w/w) ranging from 10-30% textured vegetable protein and 5-25% pea protein were recorded during a 14-day storage period. Spectral profiles extracted from VNIR-HSI were pre-processed with Savitzky?Golay 1st derivative with 11 smoothing points and mean centering. Unsupervised Principal Component Analysis (PCA) model on pre-processed data classified 140 PBMB samples based on storage days, formulations, and explored 10 wavelengths significantly contributing to predict their quality. Partial Least Square Regression (PLSR) was conducted both in the pre-processed full spectral range and selected wavelengths, to predict the quality traits. The figures of merit of PLSR models yielded good prediction of pH, redness, moisture, and hardness with low error values. The results unravel the feasibility of VNIR-HSI as a real-time and non-invasive method for predicting potential chemical changes of PBMA during storage.
Keywords: Plant-based meat, Hyperspectral imaging, Food quality, Plant proteins, Multivariate analysis
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2023-07-23
Technical field:
Conference name: CSBE/SCGAB 2023 Annual Conference, Lethbridge, Alberta, 23-26 July 2023.
Session name:
Other information:
Type: Presentation
Format:
Publication type: Text.Abstract
Source:
Relation:
Coverage: North_America
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
©
2026
CSBE-SCGAB