Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning
Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) offers an elevated signal-to-noise ratio (SNR), enabling exceptionally high spatial resolution that benefits both clinical diagnostics and advanced research. However, the jump to higher fields introduces complications, particularly transmit radi...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2025-09-01
|
| Series: | Meta-Radiology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2950162825000347 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849248162228731904 |
|---|---|
| author | Zhengyi Lu Hao Liang Ming Lu Xiao Wang Xinqiang Yan Yuankai Huo |
| author_facet | Zhengyi Lu Hao Liang Ming Lu Xiao Wang Xinqiang Yan Yuankai Huo |
| author_sort | Zhengyi Lu |
| collection | DOAJ |
| description | Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) offers an elevated signal-to-noise ratio (SNR), enabling exceptionally high spatial resolution that benefits both clinical diagnostics and advanced research. However, the jump to higher fields introduces complications, particularly transmit radiofrequency (RF) field (B1+) inhomogeneities, manifesting as uneven flip angles and image intensity irregularities. These artifacts can degrade image quality and impede broader clinical adoption. Traditional RF shimming methods, such as Magnitude Least Squares (MLS) optimization, effectively mitigate B1+ inhomogeneity, but remain time-consuming. Recent machine learning approaches, including RF Shim Prediction by Iteratively Projected Ridge Regression and other deep learning architectures, suggest alternative pathways. Although these approaches show promise, challenges such as extensive training periods, limited network complexity, and practical data requirements persist. In this paper, we introduce a holistic learning-based framework called Fast-RF-Shimming, which achieves a 5000 × speed-up compared to the traditional MLS method. In the initial phase, we employ random-initialized Adaptive Moment Estimation (Adam) to derive the desired reference shimming weights from multi-channel B1+ fields. Next, we train a Residual Network (ResNet) to map B1+ fields directly to the ultimate RF shimming outputs, incorporating the confidence parameter into its loss function. Finally, we design Non-uniformity Field Detector (NFD), an optional post-processing step, to ensure the extreme non-uniform outcomes are identified. Comparative evaluations with standard MLS optimization underscore notable gains in both processing speed and predictive accuracy, which indicates that our technique shows a promising solution for addressing persistent inhomogeneity challenges. |
| format | Article |
| id | doaj-art-389e1d6e710d45129785b0ca5e3b49fc |
| institution | Kabale University |
| issn | 2950-1628 |
| language | English |
| publishDate | 2025-09-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Meta-Radiology |
| spelling | doaj-art-389e1d6e710d45129785b0ca5e3b49fc2025-08-20T03:58:00ZengKeAi Communications Co., Ltd.Meta-Radiology2950-16282025-09-013310016610.1016/j.metrad.2025.100166Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learningZhengyi Lu0Hao Liang1Ming Lu2Xiao Wang3Xinqiang Yan4Yuankai Huo5Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USA; Corresponding author.Vanderbilt University Institute of Imaging Science, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USAVanderbilt University Institute of Imaging Science, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USAComputational Science and Engineering Division, Oak Ridge National Laboratory, Oak Ridge, TN, USAVanderbilt University Institute of Imaging Science, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USADepartment of Computer Science, Vanderbilt University, Nashville, TN, USA; Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USAUltrahigh field (UHF) Magnetic Resonance Imaging (MRI) offers an elevated signal-to-noise ratio (SNR), enabling exceptionally high spatial resolution that benefits both clinical diagnostics and advanced research. However, the jump to higher fields introduces complications, particularly transmit radiofrequency (RF) field (B1+) inhomogeneities, manifesting as uneven flip angles and image intensity irregularities. These artifacts can degrade image quality and impede broader clinical adoption. Traditional RF shimming methods, such as Magnitude Least Squares (MLS) optimization, effectively mitigate B1+ inhomogeneity, but remain time-consuming. Recent machine learning approaches, including RF Shim Prediction by Iteratively Projected Ridge Regression and other deep learning architectures, suggest alternative pathways. Although these approaches show promise, challenges such as extensive training periods, limited network complexity, and practical data requirements persist. In this paper, we introduce a holistic learning-based framework called Fast-RF-Shimming, which achieves a 5000 × speed-up compared to the traditional MLS method. In the initial phase, we employ random-initialized Adaptive Moment Estimation (Adam) to derive the desired reference shimming weights from multi-channel B1+ fields. Next, we train a Residual Network (ResNet) to map B1+ fields directly to the ultimate RF shimming outputs, incorporating the confidence parameter into its loss function. Finally, we design Non-uniformity Field Detector (NFD), an optional post-processing step, to ensure the extreme non-uniform outcomes are identified. Comparative evaluations with standard MLS optimization underscore notable gains in both processing speed and predictive accuracy, which indicates that our technique shows a promising solution for addressing persistent inhomogeneity challenges.http://www.sciencedirect.com/science/article/pii/S2950162825000347RF shimming designMagnetic field inhomogeneityDeep learning |
| spellingShingle | Zhengyi Lu Hao Liang Ming Lu Xiao Wang Xinqiang Yan Yuankai Huo Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning Meta-Radiology RF shimming design Magnetic field inhomogeneity Deep learning |
| title | Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning |
| title_full | Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning |
| title_fullStr | Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning |
| title_full_unstemmed | Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning |
| title_short | Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning |
| title_sort | fast rf shimming accelerate rf shimming in 7t mri using deep learning |
| topic | RF shimming design Magnetic field inhomogeneity Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2950162825000347 |
| work_keys_str_mv | AT zhengyilu fastrfshimmingacceleraterfshimmingin7tmriusingdeeplearning AT haoliang fastrfshimmingacceleraterfshimmingin7tmriusingdeeplearning AT minglu fastrfshimmingacceleraterfshimmingin7tmriusingdeeplearning AT xiaowang fastrfshimmingacceleraterfshimmingin7tmriusingdeeplearning AT xinqiangyan fastrfshimmingacceleraterfshimmingin7tmriusingdeeplearning AT yuankaihuo fastrfshimmingacceleraterfshimmingin7tmriusingdeeplearning |