A miniaturized NIR-based approach for quantifying fat content and cow milk adulteration in goat milk

This study introduces a novel approach for determining the fat content and cow milk adulteration in goat milk using a miniaturized NIR spectrometer coupled with multivariate calibration frameworks based on the Successive Projections Algorithm for variable and interval selection in Multiple Linear Regression (SPA-MLR) and Partial Least Squares (iSPA-PLS). An 11-point Savitzky-Golay smoothing (SGS) demonstrated the best predictive performance among the preprocessing techniques. The SGS/iSPA-PLS model achieved correlation coefficients (r_pred) of 0.97 and 0.99, root mean square errors of prediction (RMSEP) of 0.12 g/100 g and 2.15 g/100 g, ratios of performance to deviation (RPD) of 4.32 and 8.96, and relative errors of prediction (REP) of 2.70 % and 8.04 % for the fat content estimation and cow milk adulteration detection, respectively. This methodology addresses key challenges in compositional variability and adulteration, offering a robust tool for advancing goat milk quality control in both research and industrial settings.