In the medical domain, acquiring large datasets is challenging due to both accessibility issues and stringent privacy regulations. Consequently, data availability and privacy protection are major obstacles to applying machine learning in medical imaging. To address this, our study proposes the Med-LSDM (Latent Semantic Diffusion Model), which operates directly in the 3D domain and leverages de-identified semantic maps to generate synthetic data as a method of privacy preservation and data augmentation. Unlike many existing methods that focus on generating 2D slices, Med-LSDM is designed specifically for 3D semantic image synthesis, making it well-suited for applications requiring full volumetric data. Med-LSDM incorporates a guiding mechanism that controls the 3D image generation process by applying a diffusion model within the latent space of a pre-trained VQ-GAN. By operating in the compressed latent space, the model significantly reduces computational complexity while still preserving critical 3D spatial details. Our approach demonstrates strong performance in 3D semantic medical image synthesis, achieving a 3D-FID score of 0.0054 on the conditional Duke Breast dataset and similar Dice scores (0.70964) to those of real images (0.71496). These results demonstrate that the synthetic data from our model have a small domain gap with real data and are useful for data augmentation.

Accurate segmentation of hepatic and portal veins is critical for preoperative planning in liver surgery, especially for resection and transplantation procedures. Extensive anatomical variability, pathological alterations, and inherent class imbalance between background and vascular structures challenge this task. Current state-of-the-art deep learning approaches often fail to generalize across patient variability or maintain vascular topology, thus limiting their clinical applicability. To overcome these limitations, we propose the D²-RD-UNet, a dual-stage, dual-class segmentation framework for hepatic and portal vessels. The D²-RD-UNet architecture employs dense and residual connections to improve feature propagation and segmentation accuracy. Our D²-RD-UNet integrates advanced data-driven preprocessing, a dual-path architecture for 3D and 4D data, with the latter concatenating computed tomography (CT) scans with four relevant vesselness filters (Sato, Frangi, OOF, and RORPO). The pipeline is completed by the first developed postprocessing multi-class vessel connectivity correction algorithm based on centerlines. Additionally, we introduce the first radius-based branching algorithm to evaluate the model's predictions locally, providing detailed insights into the accuracy of vascular reconstructions at different scales. In order to make up for the scarcity of well-annotated open datasets for hepatic vessels segmentation, we curated AIMS-HPV-385, a large, pathological, multi-class, and validated dataset on 385 CT scans. We trained different configurations of D²-RD-UNet and state-of-the-art models on 327 CTs of AIMS-HPV-385. Experimental results on the remaining 58 CTs of AIMS-HPV-385 and on the 20 CTs of 3D-IRCADb-01 demonstrate superior performances of the D²-RD-UNet variants over state-of-the-art methods, achieving robust generalization, preserving vascular continuity, and offering a reliable approach for liver vascular reconstructions.

Spiculation, characterized by irregular, spike-like projections from nodule margins, serves as a crucial radiological biomarker for malignancy assessment and early cancer detection. These distinctive stellate patterns strongly correlate with tumor invasiveness and are vital for accurate diagnosis and treatment planning. Traditional computer-aided diagnosis (CAD) systems are limited in their capability to capture and use these patterns given their subtlety, difficulty in quantifying them, and small datasets available to learn these patterns. To address these challenges, we propose a novel framework leveraging variational autoencoders (VAE) to discover, extract, and vary disentangled latent representations of lung nodule images. By gradually varying the latent representations of non-spiculated nodule images, we generate augmented datasets containing spiculated nodule variations that, we hypothesize, can improve the diagnostic classification of lung nodules. Using the National Institutes of Health/National Cancer Institute Lung Image Database Consortium (LIDC) dataset, our results show that incorporating these spiculated image variations into the classification pipeline significantly improves spiculation detection performance up to 7.53%. Notably, this enhancement in spiculation detection is achieved while preserving the classification performance of non-spiculated cases. This approach effectively addresses class imbalance and enhances overall classification outcomes. The gradual attenuation of spiculation characteristics demonstrates our model's ability to both capture and generate clinically relevant semantic features in an algorithmic manner. These findings suggest that the integration of semantic-based latent representations into CAD models not only enhances diagnostic accuracy but also provides insights into the underlying morphological progression of spiculated nodules, enabling more informed and clinically meaningful AI-driven support systems.

This study introduces a novel evaluation framework, GPTRadScore, to systematically assess the performance of multimodal large language models (MLLMs) in generating clinically accurate findings from CT imaging. Specifically, GPTRadScore leverages LLMs as an evaluation metric, aiming to provide a more accurate and clinically informed assessment than traditional language-specific methods. Using this framework, we evaluate the capability of several MLLMs, including GPT-4 with Vision (GPT-4V), Gemini Pro Vision, LLaVA-Med, and RadFM, to interpret findings in CT scans. This retrospective study leverages a subset of the public DeepLesion dataset to evaluate the performance of several multimodal LLMs in describing findings in CT slices. GPTRadScore was developed to assess the generated descriptions (location, body part, and type) using GPT-4, alongside traditional metrics. RadFM was fine-tuned using a subset of the DeepLesion dataset with additional labeled examples targeting complex findings. Post fine-tuning, performance was reassessed using GPTRadScore to measure accuracy improvements. Evaluations demonstrated a high correlation of GPTRadScore with clinician assessments, with Pearson's correlation coefficients of 0.87, 0.91, 0.75, 0.90, and 0.89. These results highlight its superiority over traditional metrics, such as BLEU, METEOR, and ROUGE, and indicate that GPTRadScore can serve as a reliable evaluation metric. Using GPTRadScore, it was observed that while GPT-4V and Gemini Pro Vision outperformed other models, significant areas for improvement remain, primarily due to limitations in the datasets used for training. Fine-tuning RadFM resulted in substantial accuracy gains: location accuracy increased from 3.41% to 12.8%, body part accuracy improved from 29.12% to 53%, and type accuracy rose from 9.24% to 30%. These findings reinforce the hypothesis that fine-tuning RadFM can significantly enhance its performance. GPT-4 effectively correlates with expert assessments, validating its use as a reliable metric for evaluating multimodal LLMs in radiological diagnostics. Additionally, the results underscore the efficacy of fine-tuning approaches in improving the descriptive accuracy of LLM-generated medical imaging findings.

<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 evaluate an open-source deep learning model for detection and localization of breast cancer on MRI. Materials and Methods In this retrospective study, a deep learning model for breast cancer detection and localization was trained on the largest breast MRI dataset to date. Data included all breast MRIs conducted at a tertiary cancer center in the United States between 2002 and 2019. The model was validated on sagittal MRIs from the primary site (<i>n</i> = 6,615 breasts). Generalizability was assessed by evaluating model performance on axial data from the primary site (<i>n</i> = 7,058 breasts) and a second clinical site (<i>n</i> = 1,840 breasts). Results The primary site dataset included 30,672 sagittal MRI examinations (52,598 breasts) from 9,986 female patients (mean [SD] age, 53 [11] years). The model achieved an area under the receiver operating characteristic curve (AUC) of 0.95 for detecting cancer in the primary site. At 90% specificity (5717/6353), model sensitivity was 83% (217/262), which was comparable to historical performance data for radiologists. The model generalized well to axial examinations, achieving an AUC of 0.92 on data from the same clinical site and 0.92 on data from a secondary site. The model accurately located the tumor in 88.5% (232/262) of sagittal images, 92.8% (272/293) of axial images from the primary site, and 87.7% (807/920) of secondary site axial images. Conclusion The model demonstrated state-of-the-art performance on breast cancer detection. Code and weights are openly available to stimulate further development and validation. ©RSNA, 2025.

Positron Emission Tomography / Computed Tomography (PET/CT) plays a critical role in medical imaging, combining functional and anatomical information to aid in accurate diagnosis. However, image quality degradation due to noise, compression and other factors could potentially lead to diagnostic uncertainty and increase the risk of misdiagnosis. When evaluating the quality of a PET/CT image, both low-level features like distortions and high-level features like organ anatomical structures affect the diagnostic value of the image. However, existing medical image quality assessment (IQA) methods are unable to account for both feature types simultaneously. In this work, we propose MS-IQA, a novel multi-scale feature fusion network for PET/CT IQA, which utilizes multi-scale features from various intermediate layers of ResNet and Swin Transformer, enhancing its ability of perceiving both local and global information. In addition, a multi-scale feature fusion module is also introduced to effectively combine high-level and low-level information through a dynamically weighted channel attention mechanism. Finally, to fill the blank of PET/CT IQA dataset, we construct PET-CT-IQA-DS, a dataset containing 2,700 varying-quality PET/CT images with quality scores assigned by radiologists. Experiments on our dataset and the publicly available LDCTIQAC2023 dataset demonstrate that our proposed model has achieved superior performance against existing state-of-the-art methods in various IQA metrics. This work provides an accurate and efficient IQA method for PET/CT. Our code and dataset are available at https://github.com/MS-IQA/MS-IQA/.

Purpose The limited availability of bronchoscopy images makes image synthesis particularly interesting for training deep learning models. Robust image translation across different domains-virtual bronchoscopy, phantom as well as in vivo and ex vivo image data-is pivotal for clinical applications. Methods This paper proposes BronchoGAN introducing anatomical constraints for image-to-image translation being integrated into a conditional GAN. In particular, we force bronchial orifices to match across input and output images. We further propose to use foundation model-generated depth images as intermediate representation ensuring robustness across a variety of input domains establishing models with substantially less reliance on individual training datasets. Moreover, our intermediate depth image representation allows to easily construct paired image data for training. Results Our experiments showed that input images from different domains (e.g., virtual bronchoscopy, phantoms) can be successfully translated to images mimicking realistic human airway appearance. We demonstrated that anatomical settings (i.e., bronchial orifices) can be robustly preserved with our approach which is shown qualitatively and quantitatively by means of improved FID, SSIM and dice coefficients scores. Our anatomical constraints enabled an improvement in the Dice coefficient of up to 0.43 for synthetic images. Conclusion Through foundation models for intermediate depth representations and bronchial orifice segmentation integrated as anatomical constraints into conditional GANs, we are able to robustly translate images from different bronchoscopy input domains. BronchoGAN allows to incorporate public CT scan data (virtual bronchoscopy) in order to generate large-scale bronchoscopy image datasets with realistic appearance. BronchoGAN enables to bridge the gap of missing public bronchoscopy images.

Vascular diseases pose a significant threat to human health, with X-ray angiography established as the gold standard for diagnosis, allowing for detailed observation of blood vessels. However, angiographic X-rays expose personnel and patients to higher radiation levels than non-angiographic X-rays, which are unwanted. Thus, modality translation from non-angiographic to angiographic X-rays is desirable. Data-driven deep approaches are hindered by the lack of paired large-scale X-ray angiography datasets. While making high-quality vascular angiography synthesis crucial, it remains challenging. We find that current medical image synthesis primarily operates at pixel level and struggles to adapt to the complex geometric structure of blood vessels, resulting in unsatisfactory quality of blood vessel image synthesis, such as disconnections or unnatural curvatures. To overcome this issue, we propose a self-supervised method via diffusion models to transform non-angiographic X-rays into angiographic X-rays, mitigating data shortages for data-driven approaches. Our model comprises a diffusion model that learns the distribution of vascular data from diffusion latent, a generator for vessel synthesis, and a mask-based adversarial module. To enhance geometric accuracy, we propose a parametric vascular model to fit the shape and distribution of blood vessels. The proposed method contributes a pipeline and a synthetic dataset for X-ray angiography. We conducted extensive comparative and ablation experiments to evaluate the Angio-Diff. The results demonstrate that our method achieves state-of-the-art performance in synthetic angiography image quality and more accurately synthesizes the geometric structure of blood vessels. The code is available at https://github.com/zfw-cv/AngioDiff.

Computed Tomography (CT) plays a crucial role in clinical diagnosis, but the growing demand for CT examinations has raised concerns about diagnostic errors. While Multimodal Large Language Models (MLLMs) demonstrate promising comprehension of medical knowledge, their tendency to produce inaccurate information highlights the need for rigorous validation. However, existing medical visual question answering (VQA) benchmarks primarily focus on simple visual recognition tasks, lacking clinical relevance and failing to assess expert-level knowledge. We introduce MedErr-CT, a novel benchmark for evaluating medical MLLMs' ability to identify and correct errors in CT reports through a VQA framework. The benchmark includes six error categories - four vision-centric errors (Omission, Insertion, Direction, Size) and two lexical error types (Unit, Typo) - and is organized into three task levels: classification, detection, and correction. Using this benchmark, we quantitatively assess the performance of state-of-the-art 3D medical MLLMs, revealing substantial variation in their capabilities across different error types. Our benchmark contributes to the development of more reliable and clinically applicable MLLMs, ultimately helping reduce diagnostic errors and improve accuracy in clinical practice. The code and datasets are available at https://github.com/babbu3682/MedErr-CT.

Endoscopic image classification plays a pivotal role in medical diagnostics by identifying anatomical landmarks and pathological findings. However, conventional closed-set classification frameworks are inherently limited in open-world clinical settings, where previously unseen conditions can arise andcompromise model reliability. To address this, we explore the application of Open Set Recognition (OSR) techniques on the Kvasir dataset, a publicly available and diverse endoscopic image collection. In this study, we evaluate and compare the OSR capabilities of several representative deep learning architectures, including ResNet-50, Swin Transformer, and a hybrid ResNet-Transformer model, under both closed-set and open-set conditions. OpenMax is adopted as a baseline OSR method to assess the ability of these models to distinguish known classes from previously unseen categories. This work represents one of the first efforts to apply open set recognition to the Kvasir dataset and provides a foundational benchmark for evaluating OSR performance in medical image analysis. Our results offer practical insights into model behavior in clinically realistic settings and highlight the importance of OSR techniques for the safe deployment of AI systems in endoscopy.