Ultra-high resolution 7 tesla (7T) magnetic resonance imaging (MRI) provides detailed anatomical views, offering better signal-to-noise ratio, resolution and tissue contrast than 3T MRI, though at the cost of accessibility. We present an advanced deep learning model for synthesizing 7T brain MRI from 3T brain MRI. Paired 7T and 3T T1-weighted images were acquired from 172 participants (124 cognitively unimpaired, 48 impaired) from the Swedish BioFINDER-2 study. To synthesize 7T MRI from 3T images, we trained two models: a specialized U-Net, and a U-Net integrated with a generative adversarial network (GAN U-Net). Our models outperformed two additional state-of-the-art 3T-to-7T models in image-based evaluation metrics. Four blinded MRI professionals judged our synthetic 7T images as comparable in detail to real 7T images, and superior in subjective visual quality to 7T images, apparently due to the reduction of artifacts. Importantly, automated segmentations of the amygdalae of synthetic GAN U-Net 7T images were more similar to manually segmented amygdalae (n=20), than automated segmentations from the 3T images that were used to synthesize the 7T images. Finally, synthetic 7T images showed similar performance to real 3T images in downstream prediction of cognitive status using MRI derivatives (n=3,168). In all, we show that synthetic T1-weighted brain images approaching 7T quality can be generated from 3T images, which may improve image quality and segmentation, without compromising performance in downstream tasks. Future directions, possible clinical use cases, and limitations are discussed.

BackgroundFifth metatarsal (5MT) fractures are common but challenging to diagnose, particularly with limited expertise or subtle fractures. Deep learning shows promise but faces limitations due to image quality requirements. This study develops a deep learning model to detect 5MT fractures from smartphone-captured radiograph images, enhancing accessibility of diagnostic tools. MethodsA retrospective study included patients aged >18 with 5MT fractures (n=1240) and controls (n=1224). Radiographs (AP, oblique, lateral) from Electronic Health Records (EHR) were obtained and photographed using a smartphone, creating a new dataset (SP). Models using ResNet 152V2 were trained on EHR, SP, and combined datasets, then evaluated on a separate smartphone test dataset (SP-test). ResultsOn validation, the SP model achieved optimal performance (AUROC: 0.99). On the SP-test dataset, the EHR models performance decreased (AUROC: 0.83), whereas SP and combined models maintained high performance (AUROC: 0.99). ConclusionsSmartphone-specific deep learning models effectively detect 5MT fractures, suggesting their practical utility in resource-limited settings.

Lung neuroendocrine tumours (NETs, also known as carcinoids) are rapidly rising in incidence worldwide but have unknown aetiology and limited therapeutic options beyond surgery. We conducted multi-omic analyses on over 300 lung NETs including whole-genome sequencing (WGS), transcriptome profiling, methylation arrays, spatial RNA sequencing, and spatial proteomics. The integration of multi-omic data provides definitive proof of the existence of four strikingly different molecular groups that vary in patient characteristics, genomic and transcriptomic profiles, microenvironment, and morphology, as much as distinct diseases. Among these, we identify a new molecular group, enriched for highly aggressive supra-carcinoids, that displays an immune-rich microenvironment linked to tumour--macrophage crosstalk, and we uncover an undifferentiated cell population within supra-carcinoids, explaining their molecular and behavioural link to high-grade lung neuroendocrine carcinomas. Deep learning models accurately identified the Ca A1, Ca A2, and Ca B groups based on morphology alone, outperforming current histological criteria. The characteristic tumour microenvironment of supra-carcinoids and the validation of a panel of immunohistochemistry markers for the other three molecular groups demonstrates that these groups can be accurately identified based solely on morphological features, facilitating their implementation in the clinical setting. Our proposed morpho-molecular classification highlights group-specific therapeutic opportunities, including DLL3, FGFR, TERT, and BRAF inhibitors. Overall, our findings unify previously proposed molecular classifications and refine the lung cancer map by revealing novel tumour types and potential treatments, with significant implications for prognosis and treatment decision-making.

Annotation variability remains a substantial challenge in medical image segmentation, stemming from ambiguous imaging boundaries and diverse clinical expertise. Traditional deep learning methods producing single deterministic segmentation predictions often fail to capture these annotator biases. Although recent studies have explored multi-rater segmentation, existing methods typically focus on a single perspective -- either generating a probabilistic ``gold standard'' consensus or preserving expert-specific preferences -- thus struggling to provide a more omni view. In this study, we propose DiffOSeg, a two-stage diffusion-based framework, which aims to simultaneously achieve both consensus-driven (combining all experts' opinions) and preference-driven (reflecting experts' individual assessments) segmentation. Stage I establishes population consensus through a probabilistic consensus strategy, while Stage II captures expert-specific preference via adaptive prompts. Demonstrated on two public datasets (LIDC-IDRI and NPC-170), our model outperforms existing state-of-the-art methods across all evaluated metrics. Source code is available at https://github.com/string-ellipses/DiffOSeg .

Medical image synthesis presents unique challenges due to the inherent complexity and high-resolution details required in clinical contexts. Traditional generative architectures such as Generative Adversarial Networks (GANs) or Variational Auto Encoder (VAEs) have shown great promise for high-resolution image generation but struggle with preserving fine-grained details that are key for accurate diagnosis. To address this issue, we introduce Pixel Perfect MegaMed, the first vision-language foundation model to synthesize images at resolutions of 1024x1024. Our method deploys a multi-scale transformer architecture designed specifically for ultra-high resolution medical image generation, enabling the preservation of both global anatomical context and local image-level details. By leveraging vision-language alignment techniques tailored to medical terminology and imaging modalities, Pixel Perfect MegaMed bridges the gap between textual descriptions and visual representations at unprecedented resolution levels. We apply our model to the CheXpert dataset and demonstrate its ability to generate clinically faithful chest X-rays from text prompts. Beyond visual quality, these high-resolution synthetic images prove valuable for downstream tasks such as classification, showing measurable performance gains when used for data augmentation, particularly in low-data regimes. Our code is accessible through the project website - https://tehraninasab.github.io/pixelperfect-megamed.

Fetal lung maturity is a critical indicator for predicting neonatal outcomes and the need for post-natal intervention, especially for pregnancies affected by fetal growth restriction. Intra-voxel incoherent motion analysis has shown promising results for non-invasive assessment of fetal lung development, but its reliance on manual segmentation is time-consuming, thus limiting its clinical applicability. In this work, we present an automated lung maturity evaluation pipeline for diffusion-weighted magnetic resonance images that consists of a deep learning-based fetal lung segmentation model and a model-fitting lung maturity assessment. A 3D nnU-Net model was trained on manually segmented images selected from the baseline frames of 4D diffusion-weighted MRI scans. The segmentation model demonstrated robust performance, yielding a mean Dice coefficient of 82.14%. Next, voxel-wise model fitting was performed based on both the nnU-Net-predicted and manual lung segmentations to quantify IVIM parameters reflecting tissue microstructure and perfusion. The results suggested no differences between the two. Our work shows that a fully automated pipeline is possible for supporting fetal lung maturity assessment and clinical decision-making.

Accurate segmentation of orbital bones in facial computed tomography (CT) images is essential for the creation of customized implants for reconstruction of defected orbital bones, particularly challenging due to the ambiguous boundaries and thin structures such as the orbital medial wall and orbital floor. In these ambiguous regions, existing segmentation approaches often output disconnected or under-segmented results. We propose a novel framework that corrects segmentation results by leveraging consensus from multiple diffusion model outputs. Our approach employs a conditional Bernoulli diffusion model trained on diverse annotation patterns per image to generate multiple plausible segmentations, followed by a consensus-driven correction that incorporates position proximity, consensus level, and gradient direction similarity to correct challenging regions. Experimental results demonstrate that our method outperforms existing methods, significantly improving recall in ambiguous regions while preserving the continuity of thin structures. Furthermore, our method automates the manual process of segmentation result correction and can be applied to image-guided surgical planning and surgery.

Interpretability can improve the safety, transparency and trust of AI models, which is especially important in healthcare applications where decisions often carry significant consequences. Mechanistic interpretability, particularly through the use of sparse autoencoders (SAEs), offers a promising approach for uncovering human-interpretable features within large transformer-based models. In this study, we apply Matryoshka-SAE to the radiology-specialised multimodal large language model, MAIRA-2, to interpret its internal representations. Using large-scale automated interpretability of the SAE features, we identify a range of clinically relevant concepts - including medical devices (e.g., line and tube placements, pacemaker presence), pathologies such as pleural effusion and cardiomegaly, longitudinal changes and textual features. We further examine the influence of these features on model behaviour through steering, demonstrating directional control over generations with mixed success. Our results reveal practical and methodological challenges, yet they offer initial insights into the internal concepts learned by MAIRA-2 - marking a step toward deeper mechanistic understanding and interpretability of a radiology-adapted multimodal large language model, and paving the way for improved model transparency. We release the trained SAEs and interpretations: https://huggingface.co/microsoft/maira-2-sae.

Traditional echocardiographic parameters such as ejection fraction (EF) and global longitudinal strain (GLS) have limitations in the early detection of cardiac dysfunction. EF often remains normal despite underlying pathology, and GLS is influenced by load conditions and vendor variability. There is a growing need for reproducible, interpretable, and operator-independent parameters that capture subtle and global cardiac functional alterations. We introduce the Acoustic Index, a novel AI-derived echocardiographic parameter designed to quantify cardiac dysfunction from standard ultrasound views. The model combines Extended Dynamic Mode Decomposition (EDMD) based on Koopman operator theory with a hybrid neural network that incorporates clinical metadata. Spatiotemporal dynamics are extracted from echocardiographic sequences to identify coherent motion patterns. These are weighted via attention mechanisms and fused with clinical data using manifold learning, resulting in a continuous score from 0 (low risk) to 1 (high risk). In a prospective cohort of 736 patients, encompassing various cardiac pathologies and normal controls, the Acoustic Index achieved an area under the curve (AUC) of 0.89 in an independent test set. Cross-validation across five folds confirmed the robustness of the model, showing that both sensitivity and specificity exceeded 0.8 when evaluated on independent data. Threshold-based analysis demonstrated stable trade-offs between sensitivity and specificity, with optimal discrimination near this threshold. The Acoustic Index represents a physics-informed, interpretable AI biomarker for cardiac function. It shows promise as a scalable, vendor-independent tool for early detection, triage, and longitudinal monitoring. Future directions include external validation, longitudinal studies, and adaptation to disease-specific classifiers.

Deep learning has revolutionized neuroimage analysis by delivering unprecedented speed and accuracy. However, the narrow scope of many training datasets constrains model robustness and generalizability. This challenge is particularly acute in magnetic resonance imaging (MRI), where image appearance varies widely across pulse sequences and scanner hardware. A recent domain-randomization strategy addresses the generalization problem by training deep neural networks on synthetic images with randomized intensities and anatomical content. By generating diverse data from anatomical segmentation maps, the approach enables models to accurately process image types unseen during training, without retraining or fine-tuning. It has demonstrated effectiveness across modalities including MRI, computed tomography, positron emission tomography, and optical coherence tomography, as well as beyond neuroimaging in ultrasound, electron and fluorescence microscopy, and X-ray microtomography. This tutorial paper reviews the principles, implementation, and potential of the synthesis-driven training paradigm. It highlights key benefits, such as improved generalization and resistance to overfitting, while discussing trade-offs such as increased computational demands. Finally, the article explores practical considerations for adopting the technique, aiming to accelerate the development of generalizable tools that make deep learning more accessible to domain experts without extensive computational resources or machine learning knowledge.