Malte Thode: Generation of SAR values using machine learning models based on simulation data

Generation of SAR values using machine learning models based on simulation data

Funding: Freie und Hansestadt Hamburg
Contact: Malte Thode, M.Sc
01.04.2026 – 30.03.2030

Accurate determination of the specific absorption rate (SAR) is essential for ensuring the electromagnetic compatibility of devices that are in close proximity with the human body.

Full-wave simulators are an established tool for analysing the behaviour of electromagnetic waves and, consequently, the SAR. Particularly in the field of bioelectromagnetics, they enable the generation of detailed data in biological tissue, as experimental measurements there are often complex or limited. However, such simulations are highly computationally intensive and time-consuming.

The aim of this project is to develop physics-based machine learning models trained on extensive simulation data. These models should be capable of generating SAR values for various scenarios quickly and efficiently. Therefore, the data generation takes into account variations in the angle of incidence and the polarisation of the incident wave, uncertainties in the electrical properties of the tissue, and frequencies up to 1 GHz.”