Durum wheat adulteration detection by NIR spectroscopy multivariate calibration

Area of Science:

• Agricultural Science
• Analytical Chemistry
• Food Science

Background:

• Durum wheat flour is susceptible to adulteration with common bread wheat flour.
• Accurate quantification of adulteration is crucial for food quality and regulatory compliance.
• Existing methods for detecting adulteration may lack precision or efficiency.

Purpose of the Study:

• To investigate near-infrared (NIR) spectroscopy for quantifying durum wheat flour adulteration.
• To compare the performance of Partial Least Squares (PLS) and a novel wavelet-based algorithm (WILMA).
• To assess the accuracy and model complexity of NIR-based methods against official standards.

Main Methods:

• Near-infrared (NIR) spectroscopy was employed to analyze wheat flour samples.
• Multivariate calibration techniques, including PLS and the WILMA algorithm, were utilized.
• Sample adulteration levels were systematically varied to build and test calibration models.

Main Results:

• Both PLS and WILMA demonstrated satisfactory performance in quantifying adulteration.
• The WILMA algorithm achieved lower quantification uncertainty than the official Italian method.
• WILMA's feature selection capability eliminated the need for signal pretreatment, resulting in parsimonious models.

Conclusions:

• NIR spectroscopy is a viable technique for detecting and quantifying wheat flour adulteration.
• The WILMA algorithm offers a more accurate and efficient approach compared to traditional methods.
• The developed NIR method, particularly with WILMA, provides a robust tool for quality control in the milling industry.