A collection of tools and reproducible research code for analyzing human exposure to electromagnetic fields, developed during my PhD research.

Overview

This project tackles the challenge of assessing how radio-frequency electromagnetic (EM) fields (6 GHz to 300 GHz) interact with the human body. As wireless technology moves into millimeter-wave frequencies, understanding exposure becomes increasingly important.

dosipy

The core of this project is dosipy, a Python package for high-frequency EM and thermal dosimetry simulation. It includes:

  • Data models for anatomical structures
  • Numerical integration utilities for complex 3D surfaces
  • A bio-heat equation solver for thermal analysis
  • Automatic differentiation for EM field assessment

Research Applications

The playground directory contains code that reproduces results from several peer-reviewed publications:

  • Spatial averaging of absorbed power density on realistic body models
  • Incident power density assessment on anatomical human models
  • Machine learning-assisted antenna modeling
  • Exposure scenarios involving wearable devices