Optimizing Transmit Field Inhomogeneity of Parallel RF Transmit Design in 7T MRI using Deep Learning

TL;DR

The paper tackles B1+ inhomogeneity in 7T MRI by introducing a two-step deep-learning approach that predicts RF shimming weights from multi-channel B1+ maps, avoiding the time-consuming, subject-dependent MLS optimization. It combines random-initialized Adaptive Moment Estimation to obtain reference weights with a ResNet-based model that learns residual mappings from B1+ inputs to RF shimming outputs, enabling fast, per-slice predictions. Compared with MLS, the proposed method yields lower RMSE across multiple folds and drastically reduces inference time, demonstrating a practical path to improved image quality at ultrahigh fields. The approach has potential to generalize to broader RF shimming and parallel transmission applications in medical imaging.

Abstract

Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) provides a higher signal-to-noise ratio and, thereby, higher spatial resolution. However, UHF MRI introduces challenges such as transmit radiofrequency (RF) field (B1+) inhomogeneities, leading to uneven flip angles and image intensity anomalies. These issues can significantly degrade imaging quality and its medical applications. This study addresses B1+ field homogeneity through a novel deep learning-based strategy. Traditional methods like Magnitude Least Squares (MLS) optimization have been effective but are time-consuming and dependent on the patient's presence. Recent machine learning approaches, such as RF Shim Prediction by Iteratively Projected Ridge Regression and deep learning frameworks, have shown promise but face limitations like extensive training times and oversimplified architectures. We propose a two-step deep learning strategy. First, we obtain the desired reference RF shimming weights from multi-channel B1+ fields using random-initialized Adaptive Moment Estimation. Then, we employ Residual Networks (ResNets) to train a model that maps B1+ fields to target RF shimming outputs. Our approach does not rely on pre-calculated reference optimizations for the testing process and efficiently learns residual functions. Comparative studies with traditional MLS optimization demonstrate our method's advantages in terms of speed and accuracy. The proposed strategy achieves a faster and more efficient RF shimming design, significantly improving imaging quality at UHF. This advancement holds potential for broader applications in medical imaging and diagnostics.

Figures (3)

• Figure 1: The $B_{1}^{+}$ field inhomogeneity problem addressed in RF shimming design and main advantages of our proposed strategy over conventional MLS method. The image to the right presents the 8-coil $B_{1}^{+}$ data we started with. The two central blocks show our method and conventional MLS method with costed runtime over one subject which is estimated from testing prediction. On the right are two optimization results on the same slice picked from testing cases, which shows a contrast of uniformity over the $B_{1}^{+}$ field between the two methods.
• Figure 2: The procedures of training and prediction of our proposed optimization strategy. The strategy starts with simulations to get the desired $B_{1}^{+}$ as inputs of the Residual Network (ResNet18). Data augmentation is then applied. Adaptive Moment Estimation 17Kingma2014 is used to calculate the reference weights of coils as targets in training. At last, predictions are made by applying the testing data into the trained DL model.
• Figure 3: The RMSE results [% of target FA] of MLS method and proposed method are shown in the box-plots. Five folds of testing results are drawn independently in five groups in the figure. A significant difference with $p < 0.001$ is ensured.