Chunyi Liu

Chunyi Liu

Auto-generated profile, based on publicly available information

Chunyi Liu

Medical School, Nanjing University, Nanjing, China

Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 December 2021 in IEEE Transactions on Cybernetics
Verified
Journal Article
Closed access
HGHongping Gan
Hongping Gan
Yang GaoYang Gao
Yang Gao
CLChunyi Liu
Chunyi Liu
HCHaiwei Chen
Haiwei Chen
Tao ZhangTao Zhang
Tao Zhang
FLFeng Liu
Feng Liu
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 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
HCHaiwei Chen
Haiwei Chen
LGLei Guo
Lei Guo
MLMingyan Li
Mingyan Li
ADAurelien Destruel
Aurelien Destruel
CLChun Yi Liu
Chun Yi Liu
SCStuart Crozier
Stuart Crozier
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.

18 June 2020 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
CLChun Yi Liu
Chun Yi Liu
LGLei Guo
Lei Guo
MLMingyan Li
Mingyan Li
HCHaiwei Chen
Haiwei Chen
JJJin Jin
Jin Jin
SCStuart Crozier
Stuart Crozier
Magnetic resonance electrical properties tomography (MR-EPT) maps the spatial distribution of the patient's electrical conductivity and permittivity using the measured B_1 data in a magnetic resonance imaging (MRI) system. Existing MR-EPT methods are usually not clinically accessible owing to their technical limits such as strong noise sensitivity. In this study, we develop a new MR-EPT method that re-expresses the involved differential equations (DEs) based on the divergence theorem. In comparison with traditional methods, the proposed method avoids the grid-wise computation of the second-order derivatives of B $_1^+$ , thereby improving the robustness against noise. Besides, for applications where the structural information can be determined in advance, EPs of a region of interest (ROI) can be calculated in a fast and efficient manner. The proposed method is firstly validated with numerical simulations, in which a three-block phantom and an anatomically accurate Duke Head model are used to evaluate the proposed method. Experiments on the 9.4T MRI system were then conducted to validate the simulations. Both results indicated that the proposed MR-EPT solution could provide a more robust reconstruction of electrical properties maps compared with conventional methods.

5 November 2018 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
LGLei Guo
Lei Guo
JJJin Jin
Jin Jin
MLMingyan Li
Mingyan Li
YWYaohui Wang
Yaohui Wang
CLChun Yi Liu
Chun Yi Liu
SCStuart Crozier
Stuart Crozier
Current integral-equation (IE) based MR electrical properties tomography (EPT) methods utilize simulated incident radio-frequency (RF) fields, which are inaccurate and lead to reconstruction errors. To improve the accuracy and practicability of IE-based MR-EPT methods, a new approach is presented that obtains the incident fields using reference subjects and RF field mapping techniques. The Incident field approximation (IFA) is first demonstrated in this paper. This approximation assumes that two imaged subjects with similar coil/subject interactions will have similar incident RF fields, thus one can feed the estimation of the incident fields within the imaged subject into the calculation of those within a homogeneous subject (reference subject). This is done by measuring the total RF fields ( $B_1^+$ field) of the reference using field mapping techniques, using the known EPs of the reference subject and by rearranging Ampere's Law and the integral equations. The calculated incident RF fields are then used to reconstruct the EPs' distribution with a three-dimensional (3D) integral-based MR-EPT method. Numerical simulation results indicated that the incident RF fields obtained from the reference subject provide accurate 3D reconstruction of EPs with less than 16% root mean square error (RMSE) in noise-free scenario while the conventional IE method had more than 28% RMSE. The phantom-based experiments at 9.4T MRI system have also been conducted to evaluate the performance of the proposed method and the results indicated that the proposed method achieved desirable robustness against the noise in practical scenario with less than 21% RMSE while the conventional differential equation-based method showed worse than 37% RMSE.

16 October 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
SSShanshan Shan
Shanshan Shan
Yang GaoYang Gao
Yang Gao
DWDavid Waddington
David Waddington
HCHongli Chen
Hongli Chen
BWBrendan Whelan
Brendan Whelan
FLFeng Liu
Feng Liu
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 January 2026 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
JLJin Liu
Jin Liu
QLQing Lin
Qing Lin
ZXZhuang Xiong
Zhuang Xiong
SSShanshan Shan
Shanshan Shan
CLChunyi Liu
Chunyi Liu
Yang GaoYang Gao
Yang Gao
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., 8× 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 320×320 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.