Estimation method for the anisotropic electrical conductivity of<i>in vivo</i>human muscles and fat between 10 kHz and 1 MHz.

<i>Objective.</i>In low frequency dosimetry the variability in the electrical conductivity values assigned to body model tissues represents a major source of uncertainty. The aim of this study is to propose a method for estimating the conductivity of human anisotropic skeletal muscle and fat<i>in vivo</i>in the frequency range from 10 kHz to 1 MHz.<i>Approach.</i>A method based on bounded electrical impedance tomography was used. Bioimpedance measurements were performed on the legs of ten subjects. Anatomically realistic models of the legs were then created using magnetic resonance images. The inverse problem of the tissue conductivities was solved using the finite element method. The results were validated using resampling techniques. These findings were also used to study the effects of muscle anisotropy on magnetic field exposure.<i>Main results.</i>The estimated conductivities for anisotropic muscle were found to be in good agreement with values found in existing literature and the anisotropy was shown to decrease with increasing frequency, with the ratio of lateral to longitudinal conductivity increasing from 37% to 64%. The conductivity of fat was found to be almost a constant 0.07 S m^-1in the frequency range considered.<i>Significance.</i>The proposed method was shown to be a viable option when estimating<i>in vivo</i>conductivity of human tissue. The results can be used in numerical dosimetry calculations or as limits in future investigations studying conductivity with bioimpedance measurements.