This paper (purchase required) reports a proof-of-concept study to detect, with a point-of-use NIR scanner, the adulteration of ground almonds with apricot kernels . The authors built a classification model by preparing their own ground almond from different almonds (120 samples) purchased at local markets and then preparing blends (up to 50%, in 2% intervals) with ground apricot kernel. They collected NIR spectra using a portable and benchtop spectrometer and analyzed the data by Soft Independent Modeling of Class Analogy (SIMCA) and Conditional Entropy (CE) with machine learning algorithms to generate a classification model. They used Partial Least Square Regression (PLSR) and CE with machine learning algorithms to predict the levels of apricot kernel in ground almonds. The authors reported that both SIMCA and CE algorithms combined with spectral data obtained from the spectrometers provided very distinct clusters for pure and adulterated samples (100% accuracy). Both units also performed well in predicting apricot kernels using PLSR with rval>0.96 with a standard error prediction (SEP) 3.98%. They conclude that, based on the SIMCA, PLSR, and CE-based models, NIR spectroscopy showed great potential for real-time surveillance to detect apricot kernel adulteration.
Photo by Marcia Cripps on Unsplash
Comments