Chunyi Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Chunyi Liu has a total of 75 co-authors. Here are the top co-authors by number of shared publications:
- Feng Liu
University of Queensland
7 shared publications - Hongping Gan
School of Software, Northwestern Polytechnical University (NPU), Xian, China
5 shared publications - Lei Guo
School of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD, Australia
5 shared publications - Mingyan Li
School of Information Technology and Electrical Engineering, The University of Queensland, Brisbane, QLD, Australia
4 shared publications - Shanshan Shan
State Key Laboratory of Radiation Medicine and Protection, School of Radiation Medicine and Protection, Collaborative Innovation Center of Radiological Medicine of Jiangsu Higher Education Institutions, Soochow University, Suzhou, China
4 shared publications - Yang Gao
School of Computer Science and Engineering Central South University Changsha China
4 shared publications
Latest Publications
- 3 June 2026
- Published by Elsevier in Signal Processing
- Vol. 249
Parallel multi-stream dual-aggregation unfolding paradigm for compressive sensing reconstruction
- 3 June 2026
- Published by Elsevier in Signal Processing
- Vol. 249
Parallel multi-stream dual-aggregation unfolding paradigm for compressive sensing reconstruction
- 7 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 25  (6) , 2431-2435
Large-Angle Leaky-Wave Antenna for Body Magnetic Resonance Imaging at 7 T
This study presents a large-angle leaky-wave antenna (LWA) array engineered to enhance Bââş homogeneity and longitudinal coverage in the human body MRI at 7T. By increasing the wave propagation constant through optimized slot geometry and high-permittivity spacer, the proposed design generates oblique leaky-wave propagation that suppresses standing-wave (SW) artifacts. Parametric studies and full-wave simulations show that the large-angle LWA achieves nearly a two-fold increase in longitudinal Bââş coverage compared with canonical small-angle LWAs and conventional dipole antennas. In the transverse plane, it reduces normalized Bââş inhomogeneity by 26.2% relative to a dipole array, with only a modest 6.2% increase in peak 10g-averaged specific absorption rate (SAR10g). Three four-channel antenna arrays were fabricated, and 7T experimental Bââş maps corroborate the simulation results, demonstrating extended field penetration and substantially improved uniformity. These results establish the large-angle LWA as an effective and practical solution for achieving more homogeneous and spatially extensive excitation in ultra-high-field body MRI.
- 7 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 25  (6) , 2431-2435
Large-Angle Leaky-Wave Antenna for Body Magnetic Resonance Imaging at 7 T
This study presents a large-angle leaky-wave antenna (LWA) array engineered to enhance Bââş homogeneity and longitudinal coverage in the human body MRI at 7T. By increasing the wave propagation constant through optimized slot geometry and high-permittivity spacer, the proposed design generates oblique leaky-wave propagation that suppresses standing-wave (SW) artifacts. Parametric studies and full-wave simulations show that the large-angle LWA achieves nearly a two-fold increase in longitudinal Bââş coverage compared with canonical small-angle LWAs and conventional dipole antennas. In the transverse plane, it reduces normalized Bââş inhomogeneity by 26.2% relative to a dipole array, with only a modest 6.2% increase in peak 10g-averaged specific absorption rate (SAR10g). Three four-channel antenna arrays were fabricated, and 7T experimental Bââş maps corroborate the simulation results, demonstrating extended field penetration and substantially improved uniformity. These results establish the large-angle LWA as an effective and practical solution for achieving more homogeneous and spatially extensive excitation in ultra-high-field body MRI.
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75  (00189456) , 1-14
PARNet: A High-Quality Reconstruction Algorithm for Single-Harmonic Projection Magnetic Particle Imaging
Magnetic particle imaging (MPI) is an emerging quantitative imaging modality that exploits the nonlinear magnetization response of superparamagnetic nanoparticles, offering high sensitivity and real-time visualization for biomedical applications. Field-free line (FFL) scanning improves signal-to-noise ratio (SNR) and acquisition efficiency but introduces severe challenges in image reconstruction. Conventional system matrix (SM)-based reconstruction methods achieve high accuracy but suffer from extensive calibration burdens, while X-space methods enable rapid reconstruction yet remain vulnerable to FFL nonuniformity and information loss, leading to blurring and artifacts. To address these limitations, this work proposes a progressive artifact-removal network (PARNet) for FFL-based MPI reconstruction. The proposed algorithm integrates stage-wise artifact suppression, multidimensional attention modeling, and a noise-aware mechanism to progressively eliminate complex artifacts while preserving fine structural details. In support of this framework, a simulation platform with parameters consistent with our in-house MPI prototype is developed to generate large-scale, diverse datasets for efficient network training, eliminating the need for time-consuming SM calibration and mitigating reconstruction errors caused by the nonuniformity of magnetic fields along the FFL direction and filtered backprojection limitations. Comprehensive evaluations on both simulated and experimental data demonstrate that the proposed method outperforms existing approaches such as Deconv and SwinIR, achieving superior reconstruction quality with reduced blur and artifacts, accurate structural preservation, and efficient recovery from voltage signals to quantitative concentration images.
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75  (00189456) , 1-14
PARNet: A High-Quality Reconstruction Algorithm for Single-Harmonic Projection Magnetic Particle Imaging
Magnetic particle imaging (MPI) is an emerging quantitative imaging modality that exploits the nonlinear magnetization response of superparamagnetic nanoparticles, offering high sensitivity and real-time visualization for biomedical applications. Field-free line (FFL) scanning improves signal-to-noise ratio (SNR) and acquisition efficiency but introduces severe challenges in image reconstruction. Conventional system matrix (SM)-based reconstruction methods achieve high accuracy but suffer from extensive calibration burdens, while X-space methods enable rapid reconstruction yet remain vulnerable to FFL nonuniformity and information loss, leading to blurring and artifacts. To address these limitations, this work proposes a progressive artifact-removal network (PARNet) for FFL-based MPI reconstruction. The proposed algorithm integrates stage-wise artifact suppression, multidimensional attention modeling, and a noise-aware mechanism to progressively eliminate complex artifacts while preserving fine structural details. In support of this framework, a simulation platform with parameters consistent with our in-house MPI prototype is developed to generate large-scale, diverse datasets for efficient network training, eliminating the need for time-consuming SM calibration and mitigating reconstruction errors caused by the nonuniformity of magnetic fields along the FFL direction and filtered backprojection limitations. Comprehensive evaluations on both simulated and experimental data demonstrate that the proposed method outperforms existing approaches such as Deconv and SwinIR, achieving superior reconstruction quality with reduced blur and artifacts, accurate structural preservation, and efficient recovery from voltage signals to quantitative concentration images.
- 16 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45  (5) , 2414-2428
Highly Undersampled MRI Reconstruction via a Single Posterior Sampling of Diffusion Models
Incoherent k-space undersampling and deep learning-based reconstruction methods have shown great success in accelerating MRI. However, the performance of most previous methods will degrade dramatically under high acceleration factors, e.g., 8Ă or higher. Recently, denoising diffusion models (DM) have demonstrated promising results in solving this issue; however, one major drawback of the DM methods is the long inference time due to a dramatic number of iterative reverse posterior sampling steps. In this work, a Single Step Diffusion Model-based reconstruction framework, namely SSDM-MRI, is proposed for restoring MRI images from highly undersampled k-space. The proposed method achieves one-step reconstruction by first training a conditional DM and then iteratively distilling this model four times using an iterative selective distillation algorithm, which works synergistically with a shortcut reverse sampling strategy for model inference. Comprehensive experiments were carried out on both publicly available fastMRI brain and knee images, as well as an in-house multi-echo GRE (QSM) subject. Overall, the results showed that SSDM-MRI outperformed other methods in terms of numerical metrics (e.g., PSNR and SSIM), error maps, image fine details, and latent susceptibility information hidden in MRI phase images. In addition, the reconstruction time for a 320Ă320 brain slice of SSDM-MRI is only 0.45 second, which is only comparable to that of a simple U-net, making it a highly effective solution for MRI reconstruction tasks.
- 16 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45  (5) , 2414-2428
Highly Undersampled MRI Reconstruction via a Single Posterior Sampling of Diffusion Models
Incoherent k-space undersampling and deep learning-based reconstruction methods have shown great success in accelerating MRI. However, the performance of most previous methods will degrade dramatically under high acceleration factors, e.g., 8Ă or higher. Recently, denoising diffusion models (DM) have demonstrated promising results in solving this issue; however, one major drawback of the DM methods is the long inference time due to a dramatic number of iterative reverse posterior sampling steps. In this work, a Single Step Diffusion Model-based reconstruction framework, namely SSDM-MRI, is proposed for restoring MRI images from highly undersampled k-space. The proposed method achieves one-step reconstruction by first training a conditional DM and then iteratively distilling this model four times using an iterative selective distillation algorithm, which works synergistically with a shortcut reverse sampling strategy for model inference. Comprehensive experiments were carried out on both publicly available fastMRI brain and knee images, as well as an in-house multi-echo GRE (QSM) subject. Overall, the results showed that SSDM-MRI outperformed other methods in terms of numerical metrics (e.g., PSNR and SSIM), error maps, image fine details, and latent susceptibility information hidden in MRI phase images. In addition, the reconstruction time for a 320Ă320 brain slice of SSDM-MRI is only 0.45 second, which is only comparable to that of a simple U-net, making it a highly effective solution for MRI reconstruction tasks.
- 7 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-12
APG-Net: APG Optimization-Induced Multiscale Attention Network for MRI Reconstruction From Undersampled k-Space Measurements
Compressed sensing magnetic resonance imaging (CS-MRI) can quickly reconstruct magnetic resonance (MR) images from undersampled k-space measurements, thereby accelerating MRI reconstruction. However, existing CS-MRI methods suffer from insufficient information interaction and an inability to adequately capture image features during the iterative reconstruction process, resulting in the reconstructed images having room for improvement in detail. Inspired by the accelerated proximal gradient (APG) algorithm, we propose an APG optimization-induced deep multi-scale attention network, called APG-Net, for accelerating MRI reconstruction from k-space measurements. APG-Net transforms the traditional APG algorithm into a learnable network structure while introducing efficient feature extraction modules to effectively capture the features of the k-space data, resulting in high-quality MR image reconstruction. Specifically, it unrolls the iterative steps of the APG algorithm into a cascade block of fixed reconstruction phases, which include the iterative unfolding module (IUM) and multi-scale dual-attention block (MDAB). The proposed IUM enhances feature fusion by leveraging information interaction between iterative phases. Moreover, the customized MDAB is utilized to extract multi-scale feature information from the k-space measurements while enhancing attention to spatial and channel-related features in both dimensions, effectively boosting feature extraction capability. Extensive experiments demonstrate that our proposed APG-Net achieves remarkable performance on MRI reconstruction tasks in contrast to state-of-the-art methods. Our code is available at APG-Net.
- 7 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-12
APG-Net: APG Optimization-Induced Multiscale Attention Network for MRI Reconstruction From Undersampled k-Space Measurements
Compressed sensing magnetic resonance imaging (CS-MRI) can quickly reconstruct magnetic resonance (MR) images from undersampled k-space measurements, thereby accelerating MRI reconstruction. However, existing CS-MRI methods suffer from insufficient information interaction and an inability to adequately capture image features during the iterative reconstruction process, resulting in the reconstructed images having room for improvement in detail. Inspired by the accelerated proximal gradient (APG) algorithm, we propose an APG optimization-induced deep multi-scale attention network, called APG-Net, for accelerating MRI reconstruction from k-space measurements. APG-Net transforms the traditional APG algorithm into a learnable network structure while introducing efficient feature extraction modules to effectively capture the features of the k-space data, resulting in high-quality MR image reconstruction. Specifically, it unrolls the iterative steps of the APG algorithm into a cascade block of fixed reconstruction phases, which include the iterative unfolding module (IUM) and multi-scale dual-attention block (MDAB). The proposed IUM enhances feature fusion by leveraging information interaction between iterative phases. Moreover, the customized MDAB is utilized to extract multi-scale feature information from the k-space measurements while enhancing attention to spatial and channel-related features in both dimensions, effectively boosting feature extraction capability. Extensive experiments demonstrate that our proposed APG-Net achieves remarkable performance on MRI reconstruction tasks in contrast to state-of-the-art methods. Our code is available at APG-Net.
- 13 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 73  (5) , 2766-2779
Asymmetrical Planar Folded Dipole Antennas for Human Body MRI at 7 T
This study presents a new Asymmetrical Planar Folded Dipole (APFD) coil, tailored for 7 T human body magnetic resonance imaging (MRI). Compared to traditional fractionated dipole antennas, the APFD coil is designed to enhance transmit efficiency, homogeneity, and SAR management. Through rigorous numerical simulations and experimental validations, the APFD coil has exhibited superior transmit field uniformity and efficiency. Moreover, it has achieved a remarkable decrease in peak SAR 10g values, reducing them by up to 31%. Imaging results from in vivo torso scans demonstrate the coil's effectiveness, providing expanded imaging area coverage and consistent illumination of anatomical structures, along with good channel decoupling. These advancements are targeted at overcoming the electromagnetic challenges at radio frequencies inherently posed in ultra-high-field (UHF, ⼠7 T) MRI, presenting a viable option for safer and more efficient human body imaging. Although further research is necessary to evaluate its clinical applicability across various anatomical areas and patient demographics, the APFD coil represents a step forward in RF coil technology, opening new possibilities for improved diagnostic capabilities and patient care in UHF MRI.
- 13 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 73  (5) , 2766-2779
Asymmetrical Planar Folded Dipole Antennas for Human Body MRI at 7 T
This study presents a new Asymmetrical Planar Folded Dipole (APFD) coil, tailored for 7 T human body magnetic resonance imaging (MRI). Compared to traditional fractionated dipole antennas, the APFD coil is designed to enhance transmit efficiency, homogeneity, and SAR management. Through rigorous numerical simulations and experimental validations, the APFD coil has exhibited superior transmit field uniformity and efficiency. Moreover, it has achieved a remarkable decrease in peak SAR 10g values, reducing them by up to 31%. Imaging results from in vivo torso scans demonstrate the coil's effectiveness, providing expanded imaging area coverage and consistent illumination of anatomical structures, along with good channel decoupling. These advancements are targeted at overcoming the electromagnetic challenges at radio frequencies inherently posed in ultra-high-field (UHF, ⼠7 T) MRI, presenting a viable option for safer and more efficient human body imaging. Although further research is necessary to evaluate its clinical applicability across various anatomical areas and patient demographics, the APFD coil represents a step forward in RF coil technology, opening new possibilities for improved diagnostic capabilities and patient care in UHF MRI.
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73  (00189456) , 1-9
Image Reconstruction With Bâ Inhomogeneity Using a Deep Unrolled Network on an Open-Bore MRI-Linac
MRI-Linac systems require fast image reconstruction with high geometric fidelity to localize and track tumours for radiotherapy treatments. However, B 0 field inhomogeneity distortions and slow MR acquisition potentially limit the quality of the image guidance and tumour treatments. In this study, we develop an interpretable unrolled network, referred to as RebinNet, to reconstruct distortion-free images from B0 inhomogeneity-corrupted k-space for fast MRI-guided radiotherapy applications. RebinNet includes convolutional neural network (CNN) blocks to perform image regularizations and nonuniform fast Fourier Transform (NUFFT) modules to incorporate B0 inhomogeneity information. The RebinNet was trained on a publicly available MR dataset (3300 images) from eleven healthy volunteers for both fully sampled and subsampled acquisitions. 768 grid phantom and 12 human brain images acquired from an open-bore 1T MRI-Linac scanner were used to evaluate the performance of the proposed network. The RebinNet was compared with the conventional regularization algorithm and our recently developed UnUNet method in terms of root mean squared error (RMSE), structural similarity (SSIM), residual distortions, and computation time. Imaging results demonstrated that the RebinNet reconstructed images with lowest RMSE (0.92) at four-time acceleration for simulated brain images. The RebinNet preserved more image details and substantially increased the computational efficiency (3s, ten-fold faster) compared to the conventional regularization methods (30s), and had better generalization ability than the UnUNet method. The proposed RebinNet can achieve rapid image reconstruction and overcome the B 0 inhomogeneity distortions simultaneously, which would facilitate accurate and fast image guidance in radiotherapy treatments.
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73  (00189456) , 1-9
Image Reconstruction With Bâ Inhomogeneity Using a Deep Unrolled Network on an Open-Bore MRI-Linac
MRI-Linac systems require fast image reconstruction with high geometric fidelity to localize and track tumours for radiotherapy treatments. However, B 0 field inhomogeneity distortions and slow MR acquisition potentially limit the quality of the image guidance and tumour treatments. In this study, we develop an interpretable unrolled network, referred to as RebinNet, to reconstruct distortion-free images from B0 inhomogeneity-corrupted k-space for fast MRI-guided radiotherapy applications. RebinNet includes convolutional neural network (CNN) blocks to perform image regularizations and nonuniform fast Fourier Transform (NUFFT) modules to incorporate B0 inhomogeneity information. The RebinNet was trained on a publicly available MR dataset (3300 images) from eleven healthy volunteers for both fully sampled and subsampled acquisitions. 768 grid phantom and 12 human brain images acquired from an open-bore 1T MRI-Linac scanner were used to evaluate the performance of the proposed network. The RebinNet was compared with the conventional regularization algorithm and our recently developed UnUNet method in terms of root mean squared error (RMSE), structural similarity (SSIM), residual distortions, and computation time. Imaging results demonstrated that the RebinNet reconstructed images with lowest RMSE (0.92) at four-time acceleration for simulated brain images. The RebinNet preserved more image details and substantially increased the computational efficiency (3s, ten-fold faster) compared to the conventional regularization methods (30s), and had better generalization ability than the UnUNet method. The proposed RebinNet can achieve rapid image reconstruction and overcome the B 0 inhomogeneity distortions simultaneously, which would facilitate accurate and fast image guidance in radiotherapy treatments.
- 1 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 53  (4) , 2558-2571
AutoBCS: Block-Based Image Compressive Sensing With Data-Driven Acquisition and Noniterative Reconstruction
Block compressive sensing (CS) is a well-known signal acquisition and reconstruction paradigm with widespread application prospects in science, engineering, and cybernetic systems. However, state-of-the-art block-based image CS (BCS) methods generally suffer from two issues. The sparsifying domain and the sensing matrices widely used for image acquisition are not data driven and, thus, both the features of the image and the relationships among subblock images are ignored. Moreover, it requires to address a high-dimensional optimization problem with extensive computational complexity for image reconstruction. In this article, we provide a deep learning (DL) strategy for BCS, called AutoBCS, which automatically takes the prior knowledge of images into account in the acquisition step and establishes a reconstruction model for performing fast image reconstruction. More precisely, we present a learning-based sensing matrix to accomplish image acquisition, thereby capturing and preserving more image characteristics than those captured by the existing methods. In addition, we build a noniterative reconstruction network, which provides an end-to-end BCS reconstruction framework to maximize image reconstruction efficiency. Furthermore, we investigate comprehensive comparison studies with both traditional BCS approaches and newly developed DL methods. Compared with these approaches, our proposed AutoBCS can not only provide superior performance in terms of image quality metrics (SSIM and PSNR) and visual perception but also automatically benefit reconstruction speed.
- 1 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 53  (4) , 2558-2571
AutoBCS: Block-Based Image Compressive Sensing With Data-Driven Acquisition and Noniterative Reconstruction
Block compressive sensing (CS) is a well-known signal acquisition and reconstruction paradigm with widespread application prospects in science, engineering, and cybernetic systems. However, state-of-the-art block-based image CS (BCS) methods generally suffer from two issues. The sparsifying domain and the sensing matrices widely used for image acquisition are not data driven and, thus, both the features of the image and the relationships among subblock images are ignored. Moreover, it requires to address a high-dimensional optimization problem with extensive computational complexity for image reconstruction. In this article, we provide a deep learning (DL) strategy for BCS, called AutoBCS, which automatically takes the prior knowledge of images into account in the acquisition step and establishes a reconstruction model for performing fast image reconstruction. More precisely, we present a learning-based sensing matrix to accomplish image acquisition, thereby capturing and preserving more image characteristics than those captured by the existing methods. In addition, we build a noniterative reconstruction network, which provides an end-to-end BCS reconstruction framework to maximize image reconstruction efficiency. Furthermore, we investigate comprehensive comparison studies with both traditional BCS approaches and newly developed DL methods. Compared with these approaches, our proposed AutoBCS can not only provide superior performance in terms of image quality metrics (SSIM and PSNR) and visual perception but also automatically benefit reconstruction speed.
- 9 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Engineering
- Vol. 68  (4) , 1178-1189
Metamaterial-Inspired Radiofrequency (RF) Shield With Reduced Specific Absorption Rate (SAR) and Improved Transmit Efficiency for UHF MRI
To prevent the interferences between radiofrequency (RF) coils and other components in the magnetic resonance imaging (MRI) system such as gradient coils, it is essential to place an RF shield between the RF coils and gradient coils. However, the induced currents on conventional RF shields have negative influences on the RF coil performance. To reduce these influences, metamaterial absorbers (MA), a class of metamaterials exhibiting nearly unity absorption rate for the incident electromagnetic fields, can be employed for the design of a novel RF shield. However, the adoption of metamaterials in MRI systems is usually problematic because of the bulkiness of the metamaterial structure. In this work, capacitors and metallic interconnectors are used to miniaturize the MA so that the unit MA cell can operate at the Larmor frequencies of 7T and 9.4T MRI and stay compact. This MA-RF shield is used to improve the transmit efficiency of RF surface coils and reduce the specific absorption rate (SAR) in the region of interest (ROI). It is successfully demonstrated by simulations and experiments that, compared with conventional RF shield structure, the transmit efficiency can be enhanced by more than 32% and the peak SAR value can be reduced by 22% using the MA-RF shield. Moreover, it is observed that the transmit field penetration is improved when the surface coil is used with the MA-RF shield. This proof-of-concept study suggests a new practical way for the utilization of metamaterials in ultra-high field MRI applications.
- 9 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Engineering
- Vol. 68  (4) , 1178-1189
Metamaterial-Inspired Radiofrequency (RF) Shield With Reduced Specific Absorption Rate (SAR) and Improved Transmit Efficiency for UHF MRI
To prevent the interferences between radiofrequency (RF) coils and other components in the magnetic resonance imaging (MRI) system such as gradient coils, it is essential to place an RF shield between the RF coils and gradient coils. However, the induced currents on conventional RF shields have negative influences on the RF coil performance. To reduce these influences, metamaterial absorbers (MA), a class of metamaterials exhibiting nearly unity absorption rate for the incident electromagnetic fields, can be employed for the design of a novel RF shield. However, the adoption of metamaterials in MRI systems is usually problematic because of the bulkiness of the metamaterial structure. In this work, capacitors and metallic interconnectors are used to miniaturize the MA so that the unit MA cell can operate at the Larmor frequencies of 7T and 9.4T MRI and stay compact. This MA-RF shield is used to improve the transmit efficiency of RF surface coils and reduce the specific absorption rate (SAR) in the region of interest (ROI). It is successfully demonstrated by simulations and experiments that, compared with conventional RF shield structure, the transmit efficiency can be enhanced by more than 32% and the peak SAR value can be reduced by 22% using the MA-RF shield. Moreover, it is observed that the transmit field penetration is improved when the surface coil is used with the MA-RF shield. This proof-of-concept study suggests a new practical way for the utilization of metamaterials in ultra-high field MRI applications.