Objective Voxel-based anatomical models play a central role in low-frequency numerical dosimetry and derivation of exposure limits. This study aims to quantify the effect of voxelization artifacts on the assessment of the induced in situ electric field in human body models exposed to low-frequency magnetic fields. Approach A high-resolution voxel-based whole-body model at two spatial resolutions was used to compute the induced electric field. A previously proposed effective conductance model was applied to reduce staircasing artifacts at the skin-air and skin-fat interfaces. Both whole-body uniform and localized non-uniform exposure scenarios were incorporated, with the magnetic fields aligned in different directions. The effect of staircasing artifact was evaluated by comparing the percentile values of spatially averaged electric field obtained with and without a mitigation method. Main results The percentile values of the averaged electric field, both with and without the staircasing artifact mitigation method, show high consistency. Good agreement was observed for the 99th to 99.999th percentile values of the averaged electric field strengths in both uniform and non-uniform exposure scenarios for different averaging methods. Significance This study provides the first systematic quantification of voxelization (staircasing) effects on percentile-based dosimetric metrics in the skin, which is a critical tissue for peripheral nerve stimulation. The findings demonstrate that staircasing-induced variability in the 99th-99.999th percentile in situ electric fields is negligible, thereby supporting the robustness of the current exposure limit derivation in international guidelines and standards. The conclusions regarding the robustness of the percentile metrics refer to numerical stability against voxelization artifacts and do not imply the general suitability of these percentiles for highly localized exposure scenarios.
Diao et al. (Mon,) studied this question.