Segmentation of lesions on CT enables automatic measurement for clinical assessment of chronic diseases (e.g., lymphoma). Integrating large language models (LLMs) into the lesion segmentation workflow offers the potential to combine imaging features with descriptions of lesion characteristics from the radiology reports. In this study, we investigate the feasibility of integrating text into the Swin-UMamba architecture for the task of lesion segmentation. The publicly available ULS23 DeepLesion dataset was used along with short-form descriptions of the findings from the reports. On the test dataset, a high Dice Score of 82% and low Hausdorff distance of 6.58 (pixels) was obtained for lesion segmentation. The proposed Text-Swin-UMamba model outperformed prior approaches: 37% improvement over the LLM-driven LanGuideMedSeg model (p < 0.001),and surpassed the purely image-based xLSTM-UNet and nnUNet models by 1.74% and 0.22%, respectively. The dataset and code can be accessed at https://github.com/ruida/LLM-Swin-UMamba

Computed tomography (CT) is one of the most widely used non-invasive imaging modalities for medical diagnosis. In clinical practice, CT images are usually acquired with large slice thicknesses due to the high cost of memory storage and operation time, resulting in an anisotropic CT volume with much lower inter-slice resolution than in-plane resolution. Since such inconsistent resolution may lead to difficulties in disease diagnosis, deep learning-based volumetric super-resolution methods have been developed to improve inter-slice resolution. Most existing methods conduct single-image super-resolution on the through-plane or synthesize intermediate slices from adjacent slices; however, the anisotropic characteristic of 3D CT volume has not been well explored. In this paper, we propose a novel cross-view texture transfer approach for CT slice interpolation by fully utilizing the anisotropic nature of 3D CT volume. Specifically, we design a unique framework that takes high-resolution in-plane texture details as a reference and transfers them to low-resolution through-plane images. To this end, we introduce a multi-reference non-local attention module that extracts meaningful features for reconstructing through-plane high-frequency details from multiple in-plane images. Through extensive experiments, we demonstrate that our method performs significantly better in CT slice interpolation than existing competing methods on public CT datasets including a real-paired benchmark, verifying the effectiveness of the proposed framework. The source code of this work is available at https://github.com/khuhm/ACVTT.

Magnetic resonance imaging (MRI) is commonly used in healthcare for its ability to generate diverse tissue contrasts without ionizing radiation. However, this flexibility complicates downstream analysis, as computational tools are often tailored to specific types of MRI and lack generalizability across the full spectrum of scans used in healthcare. Here, we introduce a versatile framework for the development and validation of AI models that can robustly process and analyze the full spectrum of scans achievable with MRI, enabling model deployment across scanner models, scan sequences, and age groups. Core to our framework is UltimateSynth, a technology that combines tissue physiology and MR physics in synthesizing realistic images across a comprehensive range of meaningful contrasts. This pan-contrast capability bolsters the AI development life cycle through efficient data labeling, generalizable model training, and thorough performance benchmarking. We showcase the effectiveness of UltimateSynth by training an off-the-shelf U-Net to generalize anatomical segmentation across any MR contrast. The U-Net yields highly robust tissue volume estimates, with variability under 4% across 150,000 unique-contrast images, 3.8% across 2,000+ low-field 0.3T scans, and 3.5% across 8,000+ images spanning the human lifespan from ages 0 to 100.

Generative artificial intelligence (AI) is rapidly transforming medical imaging by enabling capabilities such as data synthesis, image enhancement, modality translation, and spatiotemporal modeling. This review presents a comprehensive and forward-looking synthesis of recent advances in generative modeling including generative adversarial networks (GANs), variational autoencoders (VAEs), diffusion models, and emerging multimodal foundation architectures and evaluates their expanding roles across the clinical imaging continuum. We systematically examine how generative AI contributes to key stages of the imaging workflow, from acquisition and reconstruction to cross-modality synthesis, diagnostic support, and treatment planning. Emphasis is placed on both retrospective and prospective clinical scenarios, where generative models help address longstanding challenges such as data scarcity, standardization, and integration across modalities. To promote rigorous benchmarking and translational readiness, we propose a three-tiered evaluation framework encompassing pixel-level fidelity, feature-level realism, and task-level clinical relevance. We also identify critical obstacles to real-world deployment, including generalization under domain shift, hallucination risk, data privacy concerns, and regulatory hurdles. Finally, we explore the convergence of generative AI with large-scale foundation models, highlighting how this synergy may enable the next generation of scalable, reliable, and clinically integrated imaging systems. By charting technical progress and translational pathways, this review aims to guide future research and foster interdisciplinary collaboration at the intersection of AI, medicine, and biomedical engineering.

The U.S. Food and Drug Administration (FDA) plays an important role in ensuring safety and effectiveness of AI/ML-enabled devices through its regulatory processes. In recent years, there has been an increase in the number of these devices cleared by FDA. This study analyzes 104 FDA-approved ML-enabled medical devices from May 2021 to April 2023, extending previous research to provide a contemporary perspective on this evolving landscape. We examined clinical task, device task, device input and output, ML method and level of autonomy. Most approvals (n = 103) were via the 510(k) premarket notification pathway, indicating substantial equivalence to existing devices. Devices predominantly supported diagnostic tasks (n = 81). The majority of devices used imaging data (n = 99), with CT and MRI being the most common modalities. Device autonomy levels were distributed as follows: 52% assistive (requiring users to confirm or approve AI provided information or decision), 27% autonomous information, and 21% autonomous decision. The prevalence of assistive devices indicates a cautious approach to integrating ML into clinical decision-making, favoring support rather than replacement of human judgment.

Medical image synthesis is an important topic for both clinical and research applications. Recently, diffusion models have become a leading approach in this area. Despite their strengths, many existing methods struggle with (1) limited generalizability that only work for specific body regions or voxel spacings, (2) slow inference, which is a common issue for diffusion models, and (3) weak alignment with input conditions, which is a critical issue for medical imaging. MAISI, a previously proposed framework, addresses generalizability issues but still suffers from slow inference and limited condition consistency. In this work, we present MAISI-v2, the first accelerated 3D medical image synthesis framework that integrates rectified flow to enable fast and high quality generation. To further enhance condition fidelity, we introduce a novel region-specific contrastive loss to enhance the sensitivity to region of interest. Our experiments show that MAISI-v2 can achieve SOTA image quality with $33 \times$ acceleration for latent diffusion model. We also conducted a downstream segmentation experiment to show that the synthetic images can be used for data augmentation. We release our code, training details, model weights, and a GUI demo to facilitate reproducibility and promote further development within the community.

Foundation models are large models trained on big data which can be used for downstream tasks. In radiology, these models can potentially address several gaps in fairness and generalization, as they can be trained on massive datasets without labelled data and adapted to tasks requiring data with a small number of descriptions. This reduces one of the limiting bottlenecks in clinical model construction-data annotation-as these models can be trained through a variety of techniques that require little more than radiological images with or without their corresponding radiological reports. However, foundation models may be insufficient as they are affected-to a smaller extent when compared with traditional supervised learning approaches-by the same issues that lead to underperforming models, such as a lack of transparency/explainability, and biases. To address these issues, we advocate that the development of foundation models should not only be pursued but also accompanied by the development of a decentralized clinical validation and continuous training framework. This does not guarantee the resolution of the problems associated with foundation models, but it enables developers, clinicians and patients to know when, how and why models should be updated, creating a clinical AI ecosystem that is better capable of serving all stakeholders. CRITICAL RELEVANCE STATEMENT: Foundation models may mitigate issues like bias and poor generalization in radiology AI, but challenges persist. We propose a decentralized, cross-institutional framework for continuous validation and training to enhance model reliability, safety, and clinical utility. KEY POINTS: Foundation models trained on large datasets reduce annotation burdens and improve fairness and generalization in radiology. Despite improvements, they still face challenges like limited transparency, explainability, and residual biases. A decentralized, cross-institutional framework for clinical validation and continuous training can strengthen reliability and inclusivity in clinical AI.

Image registration is a fundamental technique in the analysis of longitudinal and multi-phase CT images within clinical practice. However, most existing methods are tailored for single-organ applications, limiting their generalizability to other anatomical regions. This work presents TotalRegistrator, an image registration framework capable of aligning multiple anatomical regions simultaneously using a standard UNet architecture and a novel field decomposition strategy. The model is lightweight, requiring only 11GB of GPU memory for training. To train and evaluate our method, we constructed a large-scale longitudinal dataset comprising 695 whole-body (thorax-abdomen-pelvic) paired CT scans from individual patients acquired at different time points. We benchmarked TotalRegistrator against a generic classical iterative algorithm and a recent foundation model for image registration. To further assess robustness and generalizability, we evaluated our model on three external datasets: the public thoracic and abdominal datasets from the Learn2Reg challenge, and a private multiphase abdominal dataset from a collaborating hospital. Experimental results on the in-house dataset show that the proposed approach generally surpasses baseline methods in multi-organ abdominal registration, with a slight drop in lung alignment performance. On out-of-distribution datasets, it achieved competitive results compared to leading single-organ models, despite not being fine-tuned for those tasks, demonstrating strong generalizability. The source code will be publicly available at: https://github.com/DIAGNijmegen/oncology_image_registration.git.

<i>"Just Accepted" papers have undergone full peer review and have been accepted for publication in <i>Radiology: Artificial Intelligence</i>. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content.</i> Purpose To develop and validate MRSegmentator, a retrospective cross-modality deep learning model for multiorgan segmentation of MRI scans. Materials and Methods This retrospective study trained MRSegmentator on 1,200 manually annotated UK Biobank Dixon MRI sequences (50 participants), 221 in-house abdominal MRI sequences (177 patients), and 1228 CT scans from the TotalSegmentator-CT dataset. A human-in-the-loop annotation workflow leveraged cross-modality transfer learning from an existing CT segmentation model to segment 40 anatomic structures. The model's performance was evaluated on 900 MRI sequences from 50 participants in the German National Cohort (NAKO), 60 MRI sequences from AMOS22 dataset, and 29 MRI sequences from TotalSegmentator-MRI. Reference standard manual annotations were used for comparison. Metrics to assess segmentation quality included Dice Similarity Coefficient (DSC). Statistical analyses included organ-and sequence-specific mean ± SD reporting and two-sided <i>t</i> tests for demographic effects. Results 139 participants were evaluated; demographic information was available for 70 (mean age 52.7 years ± 14.0 [SD], 36 female). Across all test datasets, MRSegmentator demonstrated high class wise DSC for well-defined organs (lungs: 0.81-0.96, heart: 0.81-0.94) and organs with anatomic variability (liver: 0.82-0.96, kidneys: 0.77-0.95). Smaller structures showed lower DSC (portal/splenic veins: 0.64-0.78, adrenal glands: 0.56-0.69). The average DSC on the external testing using NAKO data, ranged from 0.85 ± 0.08 for T2-HASTE to 0.91 ± 0.05 for in-phase sequences. The model generalized well to CT, achieving mean DSC of 0.84 ± 0.12 on AMOS CT data. Conclusion MRSegmentator accurately segmented 40 anatomic structures on MRI and generalized to CT; outperforming existing open-source tools. Published under a CC BY 4.0 license.

Positron emission tomography-computed tomography (PET-CT) is recommended for response evaluation in aggressive large B-cell lymphoma (LBCL) but cannot detect minimal residual disease (MRD). Circulating tumor DNA (ctDNA) has emerged as a promising biomarker for real-time disease monitoring. This study evaluated longitudinal ctDNA monitoring as an MRD marker in LBCL. In this prospective, single-center study, 14 newly diagnosed LBCL patients receiving first-line immunochemotherapy underwent frequent longitudinal blood sampling. A 53-gene targeted sequencing panel quantified ctDNA and evaluated its kinetics, correlating it with clinical parameters and PET-CT, including total metabolic tumor volume (TMTV) calculated using AI-based analysis via RECOMIA. Baseline ctDNA was detected in 11 out of 14 patients (79%), with a median variant allele frequency of 6.88% (interquartile range: 1.19-10.20%). ctDNA levels correlated significantly with TMTV (ρ = 0.91, p < 0.0001) and lactate dehydrogenase. Circulating tumor DNA kinetics, including after one treatment cycle, mirrored PET-CT metabolic changes and identified relapsing or refractory cases. This study demonstrates ctDNA-based MRD monitoring in LBCL using a fixed targeted assay with an analytical sensitivity of at least 10-3. The kinetics of ctDNA reflects the clinical course and PET-CT findings, underscoring its complementary potential to PET-CT.