Optical coherence tomography (OCT) has transformed clinical eye care by providing high-resolution volumetric imaging of the retina. Recently, ultrawide-field-of-view (FOV) OCT played an increasingly significant clinical role; however, most clinical OCT systems offer only a rather limited FOV. We increased the FOV of clinical OCT by volumetrically montaging multiple OCT datasets in three dimensions (3D). We performed volumetric montaging by representing the internal limiting membrane (ILM) and retinal pigment epithelium (RPE) in each volume as point clouds and using these point clouds to compute transformations that map each volume to a common reference frame. We validated our methodology using datasets from three institutions with different OCT hardware and data-acquisition procedures. Using the mean surface distance between point clouds, we found the error in montaging was less than the lateral pixel size. Our method enabled existing clinical OCT to achieve ultrawide FOV imaging without any hardware modification.

Purpose:RF-induced heating remains a major barrier to MRI access for patients with epicardial cardiac implantable electronic devices (CIEDs). Although ISO/TS 10974 Tier-3 transfer function (TF) methods are established for unbranched leads, no analogous framework exists for bifurcated leads, in which branch asymmetry and inter-branch coupling may substantially alter heating. We developed and validated a cumulative transfer function (cTF) framework to address this gap. Methods:Following ISO/TS 10974 Tier-3 formalism, we measured, calibrated, and validated cTFs for a commercial 35 cm bipolar epicardial lead at 1.5 T. The framework explicitly accounts for branch-specific response and cross-branch coupling. Validation was performed with 24 canonical lead configurations in a homogeneous phantom and, without recalibration, in a heterogeneous anthropomorphic pediatric phantom with clinically derived trajectories. A single-branch TF approximation served as a comparator. The validated cTF was applied to predict RF heating across adult and pediatric human models at multiple imaging landmarks. Results:Compared with the single-branch TF approximation, the cTF reduced prediction error by nearly 70% in the primary validation dataset. In secondary validation, the cTF maintained low error across clinically relevant trajectories and imaging landmarks. In human models, the framework revealed marked anatomy- and landmark-dependent variation in predicted heating for the tested 35 cm lead, with low predicted heating in pediatric models and substantially higher heating in selected adult chest and upper abdominal imaging scenarios. Conclusion:The cTF provides a validated framework for RF-heating assessment of bifurcated leads and substantially improves prediction accuracy over single-branch TF approximations that neglect branch coupling.

In accelerated MRI reconstruction, recovering anatomy from under-sampled, noisy data is challenging. While deep learning has shown promise, most models rely on convolutions that struggle with long-range dependencies. We introduce VTransFormer, a novel deep unrolled foundation model that integrates Vision Transformers with Channel Attention Blocks. It employs a multi-resolution feature fusion strategy, combining high-resolution spatial detail with low-resolution global context, and dynamically selects between global and local extractors based on adaptive quality evaluation. We comprehensively evaluated VTransFormer on the CMRxRecon Challenge 2025, including multi-center deployment across multiple new centers, multi-disease applications across various cardiovascular conditions, 5T high-field imaging, and pediatric imaging scenarios. Extensive experiments demonstrate significant improvements in SSIM, PSNR, and NMSE metrics across all challenging clinical deployment scenarios, establishing VTransFormer as an effective foundation model for diverse cardiac MRI reconstruction applications.

Chronic thromboembolic pulmonary hypertension (CTEPH) can lead to right ventricular (RV) ischemia and dysfunction due to chronic pulmonary artery obstruction and increased afterload. While cardiovascular MRI (CMR) enables noninvasive assessment of myocardial perfusion, its role in CTEPH remains unclear. The authors report adenosine stress perfusion CMR findings from two patients with CTEPH before and after pulmonary thromboendarterectomy (PTE). Both showed reduced biventricular perfusion before PTE; one demonstrated post-PTE improvement. Perfusion findings aligned with invasive hemodynamics, suggesting that CMR-derived myocardial perfusion reserve may serve as a valuable tool for assessing treatment response and RV pathophysiologic characteristics in CTEPH. Keywords: Cardiac, Pulmonary Arteries, Chronic Thromboembolic Pulmonary Hypertension, Pulmonary Thromboendarterectomy, Quantitative Perfusion Cardiovascular MRI, Myocardial Blood Flow, Myocardial Perfusion Reserve Supplemental material is available for this article. © RSNA, 2025.

OBJECTIVE:Highly-undersampled, dynamic MRI reconstruction, particularly in multi-coil scenarios, is a challenging inverse problem. Unrolled networks achieve state-of-the-art performance in MRI reconstruction but suffer from long training times and extensive GPU memory cost. METHODS:In this work, we propose a novel training strategy for IMplicit UNrolled NEtworks (IMUNNE) for highly-undersampled, multi-coil dynamic MRI reconstruction. It formulates the MRI reconstruction problem as an implicit fixed-point equation and leverages gradient approximation for backpropagation, enabling training of deep architectures with fixed memory cost. This study represents the first application of implicit network theory in the context of real-time cine MRI. The proposed method is evaluated using a prospectively undersampled, real-time cine dataset using radial k-space sampling, comprising balanced steady-state free precession (b-SSFP) readouts. Experiments include a hyperparameter search, head-to-head comparisons with a complex U-Net (CU-Net) and an alternating unrolled network (Alt-UN), and an analysis of robustness under noise perturbations; peak signal-to-noise ratio, structural similarity index, normalized root mean-square error, spatio-temporal entropic difference, and a blur metric were used. RESULTS:IMUNNE produced significantly and slightly better image quality compared to CU-Net and Alt-UN, respectively. Compared with Alt-UN, IMUNNE significantly reduced training and inference times, making it a promising approach for highly-accelerated, multi-coil real-time cine MRI reconstruction. CONCLUSION:IMUNNE strategy successfully applies unrolled networks to image reconstruction of highly-accelerated, real-time radial cine MRI. SIGNIFICANCE:Implicit training enables rapid, high-quality, and cost-effective CMR exams by reducing training and inference times and lowering memory cost associated with advanced reconstruction methods.

Atrial disease or myopathy is a growing concept in cardiovascular medicine, particularly in the context of atrial fibrillation, as well as amyloidosis and heart failure. Among cardiac imaging modalities, cardiovascular magnetic resonance (CMR) is particularly well suited for a comprehensive assessment of atrial myopathy, including tissue characterization and hemodynamics. The goal of this review article is to describe clinical applications and make recommendations on pulse sequences as well as imaging parameters to assess the left atrium and left atrial appendage. Furthermore, we aimed to create an overview of current and promising future emerging applications of left atrium-specific CMR pulse sequences focusing on both electrophysiologic (EP) and non-EP applications.

Cardiovascular Magnetic Resonance 2024 Conference (CMR2024) convened in London, UK, from 24 to 26 January 2024 and brought together 2705 learners and renowned cardiac imaging professionals to discuss and learn about the latest advancements. Organized by the Society for Cardiovascular Magnetic Resonance (SCMR) and the European Association of Cardiovascular Imaging (EACVI), in collaboration with the European Society of Cardiovascular Radiology (ESCR), CMR2024 was the largest international cardiac magnetic resonance conference to date. This conference underscored the collaboration between cardiologists, radiologists, scientists, and technologists by bringing together three major societies-SCMR, EACVI, and ESCR. Innovative session formats like 'Shark Tank' and 'Workflow, Innovations & Patients' facilitated expert opinion and practical experiences sharing in a 'TED-talk style'. With over 1168 abstract submissions and 75% acceptance rate, the programme featured multiple Early Career Award sessions, oral scientific sessions, oral case sessions, and rapid-fire sessions, all categorized by topic. Highlights included patient- and physician-centred imaging sessions, sharing referring physicians' and patients' insights of incremental value of cardiovascular magnetic resonance (CMR) in patient's management. The programme offered invited lectures in eight parallel tracks with three plenary and two keynote speakers. In addition, the interactive workshops and panel discussions provided a platform for knowledge exchange, support, and collaboration. A great emphasis was placed on collaboration between radiologists, cardiologists, scientists, and technologists, showcasing an ideal cardiac imaging marriage as a model for enhanced patient care around the globe. The event also featured exhibitions of the latest CMR technology and software, offering attendees a glimpse into the future cardiac imaging. CMR2024 emerged as a remarkable scientific, educational, and networking event, in-spiring attendees to learn and collaborate within the global CMR community.

Unsupervised domain adaptation (UDA) is essential for medical image segmentation, especially in cross-modality data scenarios. UDA aims to transfer knowledge from a labeled source domain to an unlabeled target domain, thereby reducing the dependency on extensive manual annotations. This paper presents DRL-STNet, a novel framework for cross-modality medical image segmentation that leverages generative adversarial networks (GANs), disentangled representation learning (DRL), and self-training (ST). Our method leverages DRL within a GAN to translate images from the source to the target modality. Then, the segmentation model is initially trained with these translated images and corresponding source labels and then fine-tuned iteratively using a combination of synthetic and real images with pseudo-labels and real labels. The proposed framework exhibits superior performance in abdominal organ segmentation on the FLARE challenge dataset, surpassing state-of-the-art methods by 11.4

Foundation models (FMs) represent a transformative advancement in artificial intelligence (AI), with growing applications in medical imaging. These models leverage self-attention mechanisms and are capable of processing multimodal data, such as images, text, audio, and video, across multiple scales. Although FMs require large datasets for initial training, they can be adapted to specific medical imaging tasks using smaller labeled datasets through techniques such as transfer learning, fine-tuning, prompt engineering, few-shot learning, and zero-shot learning, making them especially valuable in data-scarce settings. Many FMs also incorporate generative AI capabilities that support the creation of synthetic medical images to further address annotation limitations. Current applications span various imaging modalities in radiology, where FMs have shown potential to improve diagnostic accuracy and streamline workflows. However, clinical integration remains challenging due to issues such as limited interpretability, potential bias, privacy concerns, regulatory constraints, high computational costs, and domain shifts between training data and real-world clinical environments. Addressing these barriers will require coordinated efforts among technical developers, health care providers, and regulatory bodies. This review explores the evolving role of FMs and generative AI in radiology, highlighting recent research advances, clinical applications, and the key challenges that must be addressed for responsible deployment.

Purpose:To use robotic visible-light optical coherence tomography (vis-OCT) to study circumferential segmental Schlemm's canal (SC) anatomy in mice after topical pilocarpine administration. Methods:Anterior segment imaging using a robotic vis-OCT to maintain perpendicular laser illumination aimed at SC was performed. Sixteen mice were studied for repeatability testing and to study aqueous humor outflow (AHO) pathway response to topical drug. Pharmaceutical-grade pilocarpine (1%; n = 5) or control artificial tears (n = 9) were given, and vis-OCT imaging was performed before and 15 minutes after drug application. SC areas and volumes were measured circumferentially. Results:Circumferential vis-OCT provided high-resolution imaging of the AHO pathways. Segmental SC anatomy was visualized with the average cross-sectional area greatest temporal (3971 ± 328 µm2) and the least nasal (2727 ± 218 µm2; P = 0.018). After pilocarpine administration, the SC became larger (pilocarpine, 26.8 ± 5.0% vs. control, 8.9 ± 4.6% volume increase; P = 0.030). However, the pilocarpine alteration was also segmental, with a greater increase observed superior (pilocarpine, 31.6 ± 8.9% vs. control, 1.8 ± 5.7% volume increase; P = 0.023) and nasal (pilocarpine, 41.1 ± 15.3% vs. control, 13.9 ± 4.5% volume increase; P = 0.045). Conclusions:Circumferential noninvasive imaging of the AHO pathways was done in vivo. Segmental SC anatomy was seen, consistent with the known segmental nature of trabecular AHO. Segmental SC anatomical response to a muscarinic agonist was also observed. Segmental glaucoma drug response around the circumference of AHO pathways is an observation that may influence patient response to glaucoma treatments.