Faculty Bibliography

High-resolution quantitative susceptibility mapping (QSM) combined with susceptibility source decomposition provides a powerful approach for investigating the magnetic susceptibility alterations in Alzheimer's disease (AD). In this study, ex vivo three-dimensional multi-echo gradient-echo (mGRE) images of 5xFAD and

Radial MRI has been extensively investigated for decades and has become one of the most widely studied non-Cartesian sampling schemes in MRI research. Major advances in image reconstruction, free-breathing imaging, and highly-accelerated MRI have further expanded interest in radial acquisition. However, despite remarkable methodological advances and rapidly growing literature, Cartesian acquisition continues to dominate routine clinical MRI today. This observation suggests that it may be time to step back and revisit a question that is often under-discussed: why has radial MRI, despite decades of innovation and strong research enthusiasm, not evolved into a dominant clinical acquisition strategy? This Mini Review addresses this question from an image formation perspective, with the central argument that the repeated sampling of k-space center, which defines radial MRI, is inherently a double-edged sword. While this sampling scheme provides intrinsic robustness to motion and enables continuous data acquisition, flexible retrospective reconstruction, and incoherent undersampling, it also introduces a series of fundamental tradeoffs, including residual motion blurring, strong streaking artifacts from bright signal sources, reconstruction-related image smoothing, and increased sensitivity to system imperfections. Importantly, these limitations arise from the same mechanism that gives radial MRI its unique strengths and may not be fully overcome even with modern reconstruction approaches. This Mini Review further discusses why radial MRI may ultimately serve a complementary rather than universal role in clinical imaging and argues that the more important question today is not whether radial MRI works, but when and where it provides the greatest clinical value in modern clinical MRI.

The inherently slow acquisition speed of MRI makes abdominal imaging highly sensitive to respiratory motion artifacts. Since the early days of MRI, the development of respiratory motion compensation techniques has been an active research topic, and this field has seen substantial progress. Despite these advances, the majority of these techniques are not used in daily clinical practice, and motion management methods used in clinical abdominal MRI today have changed little over the past decades. This observation points to a significant gap between technical innovation and clinical translation in this area. This review is motivated by this question: why have so many motion management techniques not been adopted into routine clinical workflows? Unlike conventional survey-style reviews that focus on summarizing emerging methods, this article takes a different, and perhaps opposite, perspective to investigate why those technologically sophisticated innovations are misaligned with practical clinical needs. Specifically, we discuss the barriers behind the gap between research advances and clinical practice, clarify the clinical requirements for effective respiratory motion management in abdominal MRI, and highlight research directions with stronger relevance to routine workflows. The review begins with an overview of the clinical impact of respiratory motion in abdominal MRI, followed by a discussion of standard abdominal MRI sequences and their motion sensitivity. We then summarize current clinical strategies and advanced approaches, along with the barriers that hinder their clinical adoption. The article concludes with future directions and broader lessons learned from this translational gap, with the goal of guiding future developments toward improved clinical integration.

Deep learning has demonstrated strong potential for MRI reconstruction. However, conventional supervised learning requires high-quality, high-SNR reference data for network training, which are often difficult or impossible to obtain, particularly in low-field MRI. Self-supervised learning eliminates the need for reference training data but may suffer from degraded performance under low-SNR conditions. To address these limitations, we propose hybrid learning, a new training framework that integrates self-supervised and supervised learning for joint MRI reconstruction and denoising when only low-SNR training data are available.

Methods: Hybrid learning is implemented in two sequential stages. In the first stage, self-supervised learning is applied to fully sampled low-SNR data to generate higher-quality pseudo-references. In the second stage, these pseudo-references are then used as targets for supervised learning to reconstruct and denoise undersampled noisy data. The proposed method was evaluated in four experiments using simulated and real noisy MRI data of the breast, lung and brain across different field strengths (0.3T to 3T), sampling trajectories (Cartesian, spiral, and radial), noise levels, and undersampling ratios. 

Results: Hybrid learning consistently improved reconstruction quality relative to both supervised and self-supervised baselines under different acceleration rates, noise levels, and sampling patterns in all experiments. Compared with standard supervised learning using noisy references, it achieved up to 167.70% higher structural similarity index (SSIM), 95.41% lower normalized mean square error (NMSE), and 90.70% lower high-frequency error norm (HFEN). Compared with standard self-supervised learning, it achieved up to 23.88% higher SSIM, 60.85% lower NMSE, and 49.13% lower HFEN. 

Conclusion: Hybrid learning enables improved MRI reconstruction under low-SNR imaging conditions by jointly addressing noise and undersampling. It provides a practical solution for robust deep learning-based reconstruction and is particularly well suited for applications such as low-field MRI, where image quality is limited by reduced SNR.

This study introduces a domain-conditioned and temporally guided diffusion framework for accelerated dynamic MRI reconstruction, in which the reverse diffusion process is explicitly guided to model spatiotemporal structure in time-resolved data. The framework integrates temporal information from time-resolved dimensions, allowing for the concurrent capture of intraframe spatial features and interframe temporal dynamics in diffusion modeling. Meanwhile, it employs additional spatiotemporal and self-consistent frequency-temporal priors to guide the diffusion process, ensuring precise temporal alignment and enhancing fine image detail recovery. To facilitate a smooth diffusion process, the nonlinear conjugate gradient algorithm is utilized during the reverse diffusion steps. The proposed model was tested on two types of MRI data: Cartesian-acquired multicoil cardiac MRI and golden-angle-radial-acquired multicoil free-breathing lung MRI, across various undersampling rates. It achieved high-quality reconstructions, demonstrating improved temporal alignment and structural recovery compared with other competitive reconstruction methods, both qualitatively and quantitatively. This diffusion framework exhibited robust performance in handling both Cartesian and non-Cartesian acquisitions, effectively reconstructing dynamic datasets in cardiac and lung MRI under different imaging conditions.

PURPOSE/OBJECTIVE:To design 3D radial spiral phyllotaxis trajectories aimed at removing phase inconsistencies, improving image quality, and enhancing parametric mapping accuracy by acquiring nearly opposing spokes starting from both hemispheres in 3D radial k-space. METHODS:Two 3D radial trajectories, pole-to-pole and continuous spiral phyllotaxis, were developed and implemented on a 3T MRI scanner in a phase-cycled balanced steady-state free precession (bSSFP) and a spoiled gradient-echo (GRE) sequence. Image quality and k-space center phase variations were evaluated in a spherical phantom using the original and new radial phyllotaxis designs. T1/T2 was quantified and compared using phase-cycled bSSFP data acquired with the new radial trajectory designs, as well as the original phyllotaxis trajectory and a Cartesian trajectory as references, in both an MRI system phantom and the brains of three healthy volunteers. ECG-triggered whole-heart GRE data were acquired using the original and pole-to-pole phyllotaxis trajectories in three healthy volunteers and compared for image quality improvement. RESULTS:All 3D radial trajectory designs showed variations in the k-space center phase depending on the orientation of the readout spokes. Image quality improved when using the pole-to-pole and continuous phyllotaxis over the original trajectory. Scans using the original trajectory had higher T1/T2 estimation errors in comparison to the new trajectories and the Cartesian trajectory. The pole-to-pole and continuous trajectories improved T1/T2 maps of the brain and image quality for all cardiac images. CONCLUSION/CONCLUSIONS:Acquiring nearly opposing spokes in 3D radial trajectory designs compensates phase inconsistencies without requiring additional corrections, which improves quantitative imaging and anatomical visualizations.

PURPOSE/OBJECTIVE:To improve 0.55T T2-weighted PROPELLER lung MRI by developing a self-supervised framework for joint reconstruction and denoising. METHODS:T2-weighted 0.55T lung MRI datasets from 44 patients with prior COVID-19 infection were used. Each PROPELLER blade was split along the readout direction into two disjoint subsets: one subset for training an unrolled network, and the other for loss calculation. Following the Noise2Noise paradigm, this framework split k-space into two subsets with independent, matched noise but identical underlying signal, enabling joint reconstruction and denoising without external training references. For comparison, coil-wise Marchenko-Pastur Principal Component Analysis (MPPCA) denoising followed by parallel imaging reconstruction was performed. The reconstructed images were evaluated by two experienced chest radiologists. RESULTS:The self-supervised model generated lung images with improved clarity, better delineation of parenchymal and airway structures, and maintained high fidelity in cases with available CT references. In addition, the proposed framework also enabled further reduction of scan time by reconstructing images with adequate diagnostic quality from only half the number of blades. The reader study confirmed that the proposed method outperformed MPPCA across all categories (Wilcoxon signed-rank test, p < 0.001), with moderate inter-reader agreement (weighted Cohen's kappa = 0.55; percentage of exact and within ±1 point agreement = 91%). CONCLUSION/CONCLUSIONS:By leveraging the intrinsic data redundancy in PROPELLER sampling and extending the Noise2Noise concept, the proposed self-supervised framework enabled simultaneous reconstruction and denoising of lung images at 0.55T to address the low-SNR challenge at low-field. It holds great potential for broad use in other low-field MRI applications.

PURPOSE/OBJECTIVE: METHODS:Rep2Rep learning extends the Noise2Noise framework by training a neural network on two repeated MRI acquisitions, using one repetition as input and another as target, without requiring ground-truth data. It incorporates noise-adaptive training, enabling denoising generalization across varying noise-levels and flexible inference with any number of repetitions. Performance was evaluated on both synthetic noisy Brain MRI and 0.55T Prostate MRI data, and compared against supervised learning and Monte Carlo Stein's Unbiased Risk Estimator (MC-SURE). RESULTS:Rep2Rep learning outperforms MC-SURE on both synthetic and 0.55T MRI datasets. On synthetic Brain data, it achieved denoising quality comparable to supervised learning and surpassed MC-SURE, particularly in preserving structural details and reducing residual noise. On the 0.55T Prostate MRI data a reader study showed that Rep2Rep-denoised 2-average images outperformed 8-average noisy images. Rep2Rep demonstrated robustness to noise-level discrepancies between training and inference, supporting its practical implementation. CONCLUSION/CONCLUSIONS:Rep2Rep learning offers an effective self-supervised denoising for low-field MRI by leveraging routinely acquired multi-repetition data. Its noise-adaptivity enables generalization to different SNR regimes without clean reference images. This makes Rep2Rep learning a promising tool for improving image quality and scan efficiency in low-field MRI.

Supervised Deep-Learning (DL)-based reconstruction algorithms have shown state-of-the-art results for highly-undersampled dynamic Magnetic Resonance Imaging (MRI) reconstruction. However, the requirement of excessive high-quality ground-truth data hinders their applications due to the generalization problem. Recently, Implicit Neural Representation (INR) has emerged as a powerful DL-based tool for solving the inverse problem by characterizing the attributes of a signal as a continuous function of corresponding coordinates in an unsupervised manner. In this work, we proposed an INR-based method to improve dynamic MRI reconstruction from highly undersampled $\boldsymbol {k}$ -space data, which only takes spatiotemporal coordinates as inputs and does not require any training on external datasets or transfer-learning from prior images. Specifically, the proposed method encodes the dynamic MRI images into neural networks as an implicit function, and the weights of the network are learned from sparsely-acquired ( $\boldsymbol {k}$ , t)-space data itself only. Benefiting from the strong implicit continuity regularization of INR together with explicit regularization for low-rankness and sparsity, our proposed method outperforms the compared state-of-the-art methods at various acceleration factors. E.g., experiments on retrospective cardiac cine datasets show an improvement of 0.6-2.0 dB in PSNR for high accelerations (up to $40.8\times $ ). The high-quality and inner continuity of the images provided by INR exhibit great potential to further improve the spatiotemporal resolution of dynamic MRI. The code is available at: https://github.com/AMRI-Lab/INR_for_DynamicMRI.

Serial Magnetic Resonance Imaging (MRI) exams are often performed in clinical practice, offering shared anatomical and motion information across imaging sessions. However, existing reconstruction methods process each session independently without leveraging this valuable longitudinal information. In this work, we propose a novel concept of longitudinal dynamic MRI, which incorporates patient-specific prior images to exploit temporal correlations across sessions. This framework enables progressive acceleration of data acquisition and reduction of scan time as more imaging sessions become available. The concept is demonstrated using the 4D Golden-angle RAdial Sparse Parallel (GRASP) MRI, a state-of-the-art dynamic imaging technique. Longitudinal reconstruction is performed by concatenating multi-session time-resolved 4D GRASP datasets into an extended dynamic series, followed by a low-rank subspace-based reconstruction algorithm. A series of experiments were conducted to evaluate the feasibility and performance of the proposed method. Results show that longitudinal 4D GRASP reconstruction consistently outperforms standard single-session reconstruction in image quality, while preserving inter-session variations. The approach demonstrated robustness to changes in anatomy, imaging intervals, and body contour, highlighting its potential for improving imaging efficiency and consistency in longitudinal MRI applications. More generally, this work suggests a new context-aware imaging paradigm in which the more we see a patient, the faster we can image.