Mid-infrared (MIR) spectroscopy for the detection of cow’s milk in buffalo milk

Abstract

In Italy, buffalo mozzarella is a largely sold and consumed dairy product. The fraudulent adulteration of buffalo milk with cheaper and more available milk of other species is very frequent. In the present study, Fourier transform infrared spectroscopy (FTIR), in combination with multivariate analysis by partial least square (PLS) regression, was applied to quantitatively detect the adulteration of buffalo milk with cow milk by using a fully automatic equipment dedicated to the routine analysis of the milk composition. To enhance the heterogeneity, cow and buffalo bulk milk was collected for a period of over three years from different dairy farms. A total of 119 samples were used for the analysis to generate 17 different concentrations of buffalo-cow milk mixtures. This procedure was used to enhance variability and to properly randomize the trials. The obtained calibration model showed an R 2 ≥ 0.99 ( R 2 cal. = 0.99861; root mean square error of cross-validation [RMSEC] = 2.04; R 2 val. = 0.99803; root mean square error of prediction [RMSEP] = 2.84; root mean square error of cross-validation [RMSECV] = 2.44) suggesting that this method could be successfully applied in the routine analysis of buffalo milk composition, providing rapid screening for possible adulteration with cow’s milk at no additional cost.

Possible Alternatives: Identifying and Quantifying Adulteration in Buffalo, Goat, and Camel Milk Using Mid-Infrared Spectroscopy Combined with Modern Statistical Machine Learning Methods

“…Buffalo milk had the highest total solids, followed by camel milk and goat milk, while cow milk had the lowest. These data show the higher content of fat in buffalo milk, goat milk, and camel milk than in cow milk, and are in agreement with those reported in previous studies [6,18]. According to the spectra, a region of approximately 2900 cm −1 provided information about the fat molecules [29], with buffalo milk having the highest peak, followed by goat milk and camel milk, with cow milk having the lowest peak.…”

“…Therefore, MIRS technology may be an ideal solution for detecting and quantifying adulteration (e.g., adulteration with cow milk or water) in these milks. Several studies have also demonstrated the effectiveness and potential of MIRS in predicting adulteration in buffalo milk [7,18], goat milk [3,7], and camel milk [19,20], but few studies report the application of MIRS in predicting water adulteration in buffalo milk, goat milk, and camel milk.…”

How Chemometrics Can Fight Milk Adulteration

“…Indeed, the commercial milk already passed to technological operation (heat treatments, fat separation, homogenization); thus, it would be more representative of real fraud to mix the different types of milk before any unit operation. This is what was done by Spina et al [83], who described in detail the farmers, the breeds, and the sampling period and batches. Furthermore, they strengthened their experimental plan by planning a randomized pairing of cow and buffalo milk to obtain 17 adulteration levels.…”

“…MIR spectroscopy is also widely used coupled with PLS regression to detect and quantify adulterants in different milk samples. In several works, MIR was used to quantify the amount of cow's milk in more expensive milk types: buffalo [83,84], goat [85], and horse [86]. It was used to analyze coconut milk samples adulterated with water [87].…”

“…Model validation is frequently addressed by iterative validation procedures, such as cross-validation. In the considered papers, the most used cross-validation strategy is leave-one-out, to whatever degree it should be avoided for its over-optimistic results, especially in the case of exhaustive sampling procedures [13,75,83]. Indeed, it means that during the iterative recalculation of the model just one sample at a time is removed; this way the robustness of the model is poorly investigated.…”