Physical knowledge improves prediction of EM Fields
This work addresses the specific problem of improving electromagnetic field prediction accuracy for MRI applications, representing an incremental advance through physics-informed deep learning.
The researchers tackled the problem of predicting electromagnetic field distributions inside MRI radio-frequency coils by developing a physics-augmented 3D U-Net model that incorporates Gauss's law of magnetism into the loss function. Their U-Net Phys model significantly outperformed a standard U-Net, particularly in predicting fields within subjects, demonstrating the advantage of integrating physical constraints.
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.