Efficient Implementation of SAR Averaging Techniques in Computational Dosimetry for Highly Detailed Human Models

This paper presents an efficient implementation of different techniques used in the computation of averaged Specific Absorption Rate (SAR). The main purpose of this work is to enhance the accuracy and speed in the prediction of SAR averaged values regarding highly detailed human models, which present a highly inhomogeneous medium. Following the recommendations of IEEE and ICNIRP standards, three different averaging schemes have been considered, namely using a cubical, contiguous and spherical volume. The resulting algorithms have been implemented and integrated into the advanced simulation platform SEMCAD X.