Alvar whole-body model: impact of muscle anisotropy on computational dosimetry.

Objective.Computational electromagnetic dosimetry relies on accurate whole-body models to assess human exposure to electromagnetic fields. However, existing models lack anisotropic properties of tissues. This work addresses these limitations by introducing a whole-body model, Alvar, containing fully anisotropic skeletal muscles.Approach.The Alvar model has been constructed based on an anatomic atlas and developed specifically for computational dosimetry. Anisotropic models of skeletal muscles were created using Laplacian vector field simulations. Computational dosimetry was performed using anisotropic and isotropic versions of Alvar and six existing isotropic body models to estimate the electric fields induced due to exposure to spatially uniform magnetic and electric fields at 50 Hz.Main results.Modelling skeletal muscle anisotropy resulted in small variations in the 99th percentile induced electric field values (±5%) when the Alvar model was exposed to a magnetic field. When exposed to an external electric field, using anisotropic muscles systematically decreased the 99th percentile values by 13%. However, local differences in the induced electric field were larger (26%-29%) and even more significant inside muscle tissue (35%-39%).Significance.The results show that isotropic models are sufficient for calculating 99th percentile values when assessing whole-body exposure at power line frequencies but can lead to errors in local electric fields, especially in muscle tissue. This research contributes to the characterisation of uncertainty in low-frequency electromagnetic dosimetry, which can inform the development of human exposure limits.