Predicting oleogels properties using non-invasive spectroscopic techniques and machine learning

Area of Science:

• Food Science and Technology
• Analytical Chemistry
• Spectroscopy

Background:

• Oleogels are food additives requiring regulatory approval, making classification by characteristics, cost, and origin essential for consumer choice.
• Accurate quantification of oleogelator concentration, free fatty acid content, and oil loss is crucial for quality control and regulatory compliance.

Purpose of the Study:

• To develop and validate a non-invasive, eco-friendly, and rapid method for classifying oleogels based on different oleogelators.
• To quantify oleogelator concentration, free fatty acid content, and oil loss using spectroscopic techniques.
• To assess the performance of Principal Component Analysis (PCA), Random Forest (RF), and Partial Least Squares Regression (PLSR) models for oleogel analysis.

Main Methods:

• Utilized a colorimeter, Raman spectrometer, and two near-infrared (NIR) spectroscopes for spectral data acquisition.
• Prepared oleogels using sunflower and soybean oils with varying concentrations (1-10%) of beeswax, glycerol monostearate, and ethylcellulose.
• Applied spectral pretreatment, PCA, RF classification, and PLSR regression for data analysis and model development.

Main Results:

• Random Forest models achieved 100% accuracy in classifying oil type and oleogelator presence, and 94% accuracy in predicting oleogelator concentration.
• Partial Least Squares Regression models demonstrated high performance for predicting free fatty acid content and oil loss, with RPD > 3 and RER > 10.
• Spectroscopic instruments, particularly colorimeter and NIR, proved effective for monitoring additives and predicting key quality parameters.

Conclusions:

• Spectroscopic methods offer a promising, non-destructive approach for the quality control of oleogels.
• The developed method enables accurate classification and quantification, supporting regulatory compliance and consumer choice.
• Colorimetry and NIR spectroscopy are valuable tools for ensuring the quality and consistency of oleogel formulations.