Faculty Bibliography

BACKGROUND/UNASSIGNED:Large language models (LLMs) have shown promising abilities in text-based clinical tasks but they do not inherently interpret medical images such as knee radiographs. PURPOSE/UNASSIGNED:To develop a human-artificial intelligence interactive diagnostic approach, named radiology generative pretrained transformer (RadGPT), aimed at assisting and synergizing with human users for the interpretation of knee radiological images. MATERIALS AND METHODS/UNASSIGNED:A total of 22 512 knee roentgen ray images and reports were retrieved from Massachusetts General Hospital; 80% of these were used for model training and 10% were used for model testing and validation, respectively. Fifteen diagnostic imaging features (eg, osteoarthritis, effusion, joint space narrowing, osteophyte) were selected to label images based on their high frequency and clinical relevance in the retrieved official reports. Area under the curve scores were calculated for each feature to assess the diagnostic performance. To evaluate the quality of the generated medical text, historical clinical reports were used as the reference text. Several metrics for text generation tasks are applied, including BiLingual Evaluation Understudy, Recall-Oriented Understudy for Gisting Evaluation, Metric for Evaluation of Translation with Explicit Ordering, and Semantic Propositional Image Caption Evaluation. RESULTS/UNASSIGNED:RadGPT, in collaboration with human users, achieved area under the curve scores ranging from 0.76 for osteonecrosis to 0.91 for arthroplasty across 15 diagnostic categories for knee conditions. Compared with the baseline LLM method, RadGPT achieved higher scores, specifically 0.18 in BiLingual Evaluation Understudy score, 0.30 in Recall-Oriented Understudy for Gisting Evaluation-L, 0.10 in Metric for Evaluation of Translation with Explicit Ordering, and 0.15 in Semantic Propositional Image Caption Evaluation, which is significantly higher than the baseline LLM method, demonstrating good linguistic overlap and clinical consistency with the reference reports. CONCLUSION/UNASSIGNED:RadGPT has achieved advanced results in knee roentgen ray image feature recognition, illustrating the potential of LLMs in medical image interpretation. The study establishes a training protocol for developing artificial intelligence-assisted tools specifically focusing on the diagnosis and interpretation of knee radiological images.

MRI is widely used for the diagnosis and management of various abdominal diseases involving organs such as the liver, pancreas, and kidneys. However, one major limitation of MRI is its relatively slow imaging speed compared to other modalities. In addition, respiratory motion poses a significant challenge in abdominal MRI, often requiring patients to hold their breath multiple times during an exam. This requirement can be particularly challenging for sick, elderly, and pediatric patients, who may have reduced breath-holding capacity. As a result, rapid imaging plays an important role in routine clinical abdominal MRI exams. Accelerated data acquisition not only reduces overall exam time but also shortens breath-hold durations, thereby improving patient comfort and compliance. Over the past decade, significant advancements in rapid MRI have led to the development of various accelerated imaging techniques for routine clinical use. These methods improve abdominal MRI by enhancing imaging speed, motion compensation, and overall image quality. Integrating these techniques into clinical practice also enables new applications that were previously challenging. This paper provides a concise yet comprehensive overview of rapid imaging techniques applicable to abdominal MRI and discusses their advantages, limitations, and potential clinical applications. By the end of this review, readers are expected to learn the latest advances in accelerated abdominal MRI and explore new frontiers in this evolving field. Evidence Level: N/A Technical Efficacy: Stage 5.

White matter (WM) brain age, a neuroimaging-derived biomarker indicating WM microstructural changes, helps predict dementia and neurodegenerative disorder risks. The cumulative effect of chronic stress on WM brain aging remains unknown. In this study, we assessed cumulative stress using a multi-system composite allostatic load (AL) index based on inflammatory, anthropometric, respiratory, lipidemia, and glucose metabolism measures, and investigated its association with WM brain age gap (BAG), computed from diffusion tensor imaging data using a machine learning model, among 22 951 European ancestries aged 40 to 69 (51.40% women) from UK Biobank. Linear regression, Mendelian randomization, along with inverse probability weighting and doubly robust methods, were used to evaluate the impact of AL on WM BAG adjusting for age, sex, socioeconomic, and lifestyle behaviors. We found increasing one AL score unit significantly increased WM BAG by 0.29 years in association analysis and by 0.33 years in Mendelian analysis. The age- and sex-stratified analysis showed consistent results among participants 45-54 and 55-64 years old, with no significant sex difference. This study demonstrated that higher chronic stress was significantly associated with accelerated brain aging, highlighting the importance of stress management in reducing dementia and neurodegenerative disease risks.

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.

MRI is the most effective method for screening high-risk breast cancer patients. While current exams primarily rely on the qualitative evaluation of morphological features before and after contrast administration and less on contrast kinetic information, the latest developments in acquisition protocols aim to combine both. However, balancing between spatial and temporal resolution poses a significant challenge in dynamic MRI. Here, we propose a radial MRI reconstruction framework for Dynamic Contrast Enhanced (DCE) imaging, which offers a joint solution to existing spatial and temporal MRI limitations. It leverages a locally low-rank (LLR) subspace model to represent spatially localized dynamics based on tissue information. Our framework demonstrated substantial improvement in CNR, noise reduction and enables a flexible temporal resolution, ranging from a few seconds to 1-second, aided by a neural network, resulting in images with reduced undersampling penalties. Finally, our reconstruction framework also shows potential benefits for head and neck, and brain MRI applications, making it a viable alternative for a range of DCE-MRI exams.

PURPOSE/OBJECTIVE:To develop a highly accelerated non-contrast-enhanced 4D-MRA technique by combining stack-of-stars golden-angle radial acquisition with a modified self-calibrated low-rank subspace reconstruction. METHODS:A low-rank subspace reconstruction framework was introduced in radial 4D MRA (SUPER 4D MRA) by combining stack-of-stars golden-angle radial acquisition with control-label k-space subtraction-based low-rank subspace modeling. Radial 4D MRA data were acquired and reconstructed using the proposed technique on 12 healthy volunteers and 1 patient with steno-occlusive disease. The performance of SUPER 4D MRA was compared with two temporally constrained reconstruction methods (golden-angle radial sparse parallel [GRASP] and GRASP-Pro) at different acceleration rates in terms of image quality and delineation of blood dynamics. RESULTS:SUPER 4D MRA outperformed the other two reconstruction methods, offering superior image quality with a clear background and detailed delineation of cerebrovascular structures as well as great temporal fidelity in blood flow dynamics. SUPER 4D MRA maintained excellent performance even at higher acceleration rates. CONCLUSIONS:SUPER 4D MRA is a promising technique for highly accelerating 4D MRA acquisition without comprising both temporal fidelity and image quality.

PURPOSE/OBJECTIVE:The objective of this study was to develop a new MRI technique for non-invasive, free-breathing imaging of glycogen in the human liver using the nuclear Overhauser effect (NOE). METHODS:The proposed method, called GraspNOE-Dixon, uses a novel MRI sequence that combines steady-state saturation-transfer preparation with multi-echo golden-angle radial stack-of-stars sampling. Multi-echo acquisition enables fat/water-separated imaging for quantification of water-specific NOE. Image reconstruction is performed using the improved golden-angle radial sparse parallel imaging (GRASP-Pro) technique to exploit spatiotemporal correlations in dynamic images. To evaluate the proposed technique, imaging experiments were first performed on glycogen phantoms, followed by in vivo studies involving healthy volunteers and patients with fatty liver disease. In addition, a comparative assessment of signal changes before and after a 12-h fasting period was performed. RESULTS:Evaluation experiments on glycogen phantoms showed a robust linear correlation between the NOE signal and glycogen concentration. In vivo experiments demonstrated motion-robust NOE-weighted images, with potential for further acceleration. In subjects with varying liver fat content, the fat/water separation approach resulted in distortion-free Z-spectra, enabling the quantification of glycogen NOE. An approximately one-third reduction in the NOE signal was observed following a 12-h fasting period, consistent with a decrease in glycogen level. CONCLUSION/CONCLUSIONS:This study introduces a clinically feasible imaging technique, GraspNOE-Dixon, for free-breathing volumetric multi-echo imaging of hepatic glycogen at 3 T. The motion robust imaging technique developed here may also have applications in other body areas beyond liver imaging.

Nonlocal self-similarity within images has become an increasingly popular prior in deep-learning models. Despite their successful image restoration performance, such models remain largely uninterpretable due to their black-box construction. Our previous studies have shown that interpretable construction of a fully convolutional denoiser (CDLNet), with performance on par with state-of-the-art black-box counterparts, is achievable by unrolling a convolutional dictionary learning algorithm. In this manuscript, we seek an interpretable construction of a convolutional network with a nonlocal self-similarity prior that performs on par with black-box nonlocal models. We show that such an architecture can be effectively achieved by up-grading the

PURPOSE/OBJECTIVE:maps from fewer echoes. METHODS:mapping was evaluated in seven experiments. RESULTS:estimation. CONCLUSIONS:estimation from fewer echoes allows for increased volumetric coverage and/or higher slice resolution without prolonging total scan times.

Respiratory motion-induced image blurring and artifacts can compromise image quality in dynamic contrast-enhanced MRI (DCE-MRI) of the liver. Despite remarkable advances in respiratory motion detection and compensation in past years, these techniques have not yet seen widespread clinical adoption. The accuracy of image-based motion detection can be especially compromised in the presence of contrast enhancement and/or in situations involving deep and/or irregular breathing patterns. This work proposes a framework that combines GRASP-Pro (Golden-angle RAdial Sparse Parallel MRI with imProved performance) MRI with a new radial sampling scheme called navi-stack-of-stars for free-breathing DCE-MRI of the liver without the need for explicit respiratory motion compensation. A prototype 3D golden-angle radial sequence with a navi-stack-of-stars sampling scheme that intermittently acquires a 2D navigator was implemented. Free-breathing DCE-MRI of the liver was conducted in 24 subjects at 3T including 17 volunteers and 7 patients. GRASP-Pro reconstruction was performed with a temporal resolution of 0.34-0.45 s per volume, whereas standard GRASP reconstruction was performed with a temporal resolution of 15 s per volume. Motion compensation was not performed in all image reconstruction tasks. Liver images in different contrast phases from both GRASP and GRASP-Pro reconstructions were visually scored by two experienced abdominal radiologists for comparison. The nonparametric paired two-tailed Wilcoxon signed-rank test was used to compare image quality scores, and the Cohen's kappa coefficient was calculated to evaluate the inter-reader agreement. GRASP-Pro MRI with sub-second temporal resolution consistently received significantly higher image quality scores (P < 0.05) than standard GRASP MRI throughout all contrast enhancement phases and across all assessment categories. There was a substantial inter-reader agreement for all assessment categories (ranging from 0.67 to 0.89). The proposed technique using GRASP-Pro reconstruction with navi-stack-of-stars sampling holds great promise for free-breathing DCE-MRI of the liver without respiratory motion compensation.