Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.11703 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917958107791360 |
|---|---|
| author | Dulny, Andrzej Jabbarigargari, Farzad Hotho, Andreas Schreiber, Laura Maria Terekhov, Maxim Krause, Anna |
| author_facet | Dulny, Andrzej Jabbarigargari, Farzad Hotho, Andreas Schreiber, Laura Maria Terekhov, Maxim Krause, Anna |
| contents | We propose a 3D U-Net model to predict the spatial distribution of electromagnetic fields inside a radio-frequency (RF) coil with a subject present, using the phase, amplitude, and position of the coils, along with the density, permittivity, and conductivity of the surrounding medium as inputs. To improve accuracy, we introduce a physics-augmented variant, U-Net Phys, which incorporates Gauss's law of magnetism into the loss function using finite differences. We train our models on electromagnetic field simulations from CST Studio Suite for an eight-channel dipole array RF coil at 7T MRI. Experimental results show that U-Net Phys significantly outperforms the standard U-Net, particularly in predicting fields within the subject, demonstrating the advantage of integrating physical constraints into deep learning-based field prediction. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_11703 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Physical knowledge improves prediction of EM Fields Dulny, Andrzej Jabbarigargari, Farzad Hotho, Andreas Schreiber, Laura Maria Terekhov, Maxim Krause, Anna Machine Learning We propose a 3D U-Net model to predict the spatial distribution of electromagnetic fields inside a radio-frequency (RF) coil with a subject present, using the phase, amplitude, and position of the coils, along with the density, permittivity, and conductivity of the surrounding medium as inputs. To improve accuracy, we introduce a physics-augmented variant, U-Net Phys, which incorporates Gauss's law of magnetism into the loss function using finite differences. We train our models on electromagnetic field simulations from CST Studio Suite for an eight-channel dipole array RF coil at 7T MRI. Experimental results show that U-Net Phys significantly outperforms the standard U-Net, particularly in predicting fields within the subject, demonstrating the advantage of integrating physical constraints into deep learning-based field prediction. |
| title | Physical knowledge improves prediction of EM Fields |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2503.11703 |