Lard classification from other animal fats using dielectric spectroscopy technique

Lard adulteration in processed foods is a major public concern as it involves religion and health. Most lard discriminating works require huge lab-based equipment and complex sample preparation. The objective of the present work was to assess the feasibility of dielectric spectroscopy as a method for classification of fats from different animal sources, in particular, lard. The dielectric spectra of each animal fat were measured in the radio frequency of 100 Hz – 100 kHz at 45°C to 55°C. The fatty acid composition of each fat was studied by using data from gas chromatography mass spectrometry (GCMS) to explain the dielectric behaviour of each fat. The principal component analysis (PCA) and artificial neural network (ANN) were used to classify different animal fats based on their dielectric spectra. It was found that lard showed the highest dielectric constant spectra among other animal fats, and was mainly affected by the composition of C16 and C18 fatty acids. PCA classification plot showed clear performance in classifying different animal fats. Finally, ANN classification showed different animal fats were classified into their respective groups effectively at high accuracy of 85%. Dielectric spectroscopy, in combination with quantitative analysis, was concluded to provide rapid method to discriminate lard from other animal fats