Estimation of dry-cured ham composition using dielectric time domain reflectometry

Development of real-time, non-destructive methods to characterize dry-cured ham is of interest to the food industry. Since dielectric properties change depending on the composition of the food product studied, time domain reflectometry (TDR) could be a useful method to characterize dry-cured ham. In this study, samples with different compositions were measured with a TDR device equipped with an open-ended coaxial line sensor. Partial least square regression (PLSR) analysis was used to develop predictive models to determine salt, water and fat contents and a(w) in dry-cured ham. Results show that salt content (RMSEV=0.22%), water content (RMSEV=1.67%) and a(w) (RMSEV=0.0087) can be accurately determined, though fat content is determined with less precision (RMSEV=2.81%). Saltiness, dryness and fatness of the samples, in the studied range, did not affect the accuracy of the predictions. Developed predictive models were accurate enough to consider the TDR device as a useful tool for characterizing and classifying dry-cured ham in industry.