FT-NIR and FT-MIR spectroscopy to discriminate competitors, non compliance and compliance grated Parmigiano Reggiano cheese

In this investigation the potential of infrared spectroscopy, coupled to different statistical methods, were used to estimate the authenticity of grated Protected Denomination of Origin (PDO) Parmigiano Reggiano cheese (P-R). The feasibility of the analytical approach in the prediction of cheese authenticity without the use of wet chemistry was evaluated. A total of 400 plastic-sealed grated cheese samples classified as: compliance P-R, competitors, non-compliance P-R (defected P-R), and P-R with rind content of > 18%. PCA was conducted for an explorative spectra analysis. Soft independent modelling of class analogy (SIMCA) analysis and artificial neural networks (ANNs) were used to classify samples, according to different cheese categories. For both the spectroscopic techniques, PCA correctly discriminated compliance P-R from competitors, but not the P-R as a function of the rind percentage and months of ripening. SIMCA analysis accurately classified the compliance and competitors' P-R samples, while samples belonging to the classes of defected P-R and P-R with rind content > 18% were not accurately classified. ANN was more efficient than SIMCA in the classification of all the cheese classes. The results showed that NIR and MIR combined with different statistical approaches can be suitable for a sensitive, non-destructive, rapid and inexpensive screening of grated P-R cheese authenticity.