Rapid detection of adulteration in powdered camel milk using ATR-IR spectroscopy and chemometrics

Abstract

As global interest in camel milk rises because of its nutritional benefits, it is increasingly processed into powder to increase shelf life and transport. However, adulteration with cow milk powder poses a serious food safety issue. This study developed multivariate models using attenuated total reflectance infrared spectroscopy (ATR-IR) to detect adulteration in camel milk powder with cow milk powder. Partial least squares regression (PLSR) was applied after preprocessing, which included Savitzky–Golay (SG) smoothing (first and second derivatives) and multiplicative scatter correction (MSC), to quantify adulteration levels ranging from 0% (control) to 50% (w/w). Preprocessing techniques, including SG smoothing (first and second derivatives) and multiplicative scatter correction (MSC), enhanced model performance. Calibration achieved an R2 of ≥99.6% and RMSE <0.93, while cross-validation yielded an R2 of ≥99.3% and RMSE <1.35. External validation confirmed the model’s reliability with R2 >98% and RMSE <1.8. Deviations remained low across all adulteration levels, even at higher concentrations (≈0.76–0.90 for 40%). These results demonstrate that ATR-FTIR spectroscopy combined with PLSR offers a rapid, nondestructive, and reliable method for detecting adulteration in camel milk powder, making it a valuable tool for quality control in the dairy industry.