Sensibility analysis of the electrical impedance parameters by the Monte Carlo method

Abstract

Electrical properties of biological tissues are used to analyze its state in different practical applications, e.g., assessing the risk of cancer. The modeling of electrical properties usually is performed by circuital models or semiempirical models as Cole-Cole. Still, another option is available, the application of the generalized effective medium theory of induced polarization. In the framework of the effective medium theory, we proposed a model for biological tissues which we called MOPET. This model links physiological parameters of tissues with their electrical properties. However, it has a higher number of parameters than the Cole-Cole and circuital models. This large number of parameters is a limitation for the use of effective medium theory to describe the electrical properties of tissue and also suggests that some of these parameters have negligible effects. We used a Monte Carlo simulation to study the effect of each MOPET parameter on the electrical impedance spectra, where uniform random variables were used to simulate the variability of the parameters of this model. We found that the heterogeneity coefficient is the most sensible parameter, i.e., a variation above 1% in this parameter alters enormously the impedance spectra.