用物理定律增强神经网络，更准预测人体内射频线圈的电磁场分布。
Physical knowledge improves prediction of EM Fields
- 在损失函数中加入麦克斯韦方程，用有限差分法融合物理规律
- 在7特斯拉磁共振环境下，对8通道偶极子阵列进行仿真训练
- 物理增强模型在人体内部场强预测上显著优于普通U-Net
我们提出一种3D U-Net模型，基于射频线圈的相位、幅度、位置，以及周围介质的密度、介电常数和电导率，预测人体存在时线圈内部的电磁场空间分布。为提升精度，引入物理增强版本U-Net Phys，通过有限差分法将磁学高斯定律融入损失函数。模型在CST Studio Suite中对7特斯拉磁共振的8通道偶极子阵列进行电磁场仿真实验训练。实验结果表明，U-Net Phys显著优于标准U-Net，尤其在人体内部场强预测方面表现突出，验证了将物理约束融入深度学习在场预测中的优势。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。