Poor-quality sleep has been linked to increased dementia risk. We investigated the relationship between healthy sleep pattern and older brain age, and the extent to which this is mediated by systemic inflammation. The study included 27,500 adults from the UK Biobank (mean age 54.7 y, 54.0% female). The presence of five self-reported healthy sleep characteristics (early chronotype, 7-8 h daily sleep, no insomnia, no snoring, no excessive daytime sleepiness) were summed into a healthy sleep score (0-5 pts) and used to define three sleep patterns: healthy (≥4 pts), intermediate (2-3 pts), and poor (≤1 pt). Low-grade inflammation was estimated using the INFLA-score, a composite index of inflammatory biomarkers. After a mean follow-up of 8.9 y, brain age was estimated using a machine learning model based on 1079 brain MRI phenotypes and used to calculate brain age gap (BAG; i.e., brain age minus chronological age). Data were analysed using linear regression and generalised structural equation models. At baseline, 898 (3.3%) participants had poor sleep, 15,283 (55.6%) had intermediate sleep, and 11,319 (41.2%) had healthy sleep. Compared to healthy sleep, intermediate (β = 0.25 [0.11, 0.40], P = 0.010) and poor (β = 0.46 [0.05, 0.87], P < 0.001) sleep were associated with significantly higher BAG. In mediation analysis, INFLA-score mediated 6.81% and 10.42% of the associations between intermediate and poor sleep and higher BAG. Poor sleep health may accelerate brain ageing. This may be driven by higher levels of systemic inflammation. Alzheimerfonden; Demensfonden; Loo and Hans Osterman Foundation for Medical Research; the Knowledge Foundation; Swedish Research Council.

Acute pulmonary embolism (PE) is a life-threatening condition often diagnosed using CT pulmonary angiography (CTPA). However, CTPA is contraindicated in patients with contrast allergies or at risk for contrast-induced nephropathy. This study explores an AI-driven approach to generate synthetic contrast-enhanced images from non-contrast CT scans for accurate diagnosis of acute PE without contrast agents. This retrospective study used dual-energy and standard CT datasets from two institutions. The internal dataset included 84 patients: 41 PE-negative cases for generative model training and 43 patients (30 PE-positive) for diagnostic evaluation. An external dataset of 62 patients (26 PE-positive) was used for further validation. We developed a generative adversarial network (GAN) based on U-Net, trained on paired non-contrast and contrast-enhanced images. The model was optimized using contrast-enhanced L1-loss with hyperparameter λ to improve anatomical accuracy. A ConvNeXt-based classifier trained on the RSNA dataset (N = 7,122) generated per-slice PE probabilities, which were aggregated for patient-level prediction via a Random Forest model. Diagnostic performance was assessed using five-fold cross-validation on both internal and external datasets. The GAN achieved optimal image similarity at λ = 0.5, with the lowest mean absolute error (0.0089) and highest MS-SSIM (0.9674). PE classification yielded AUCs of 0.861 and 0.836 in the internal dataset, and 0.787 and 0.680 in the external dataset, using real and synthetic images, respectively. No statistically significant differences were observed. Our findings demonstrate that synthetic contrast CT can serve as a viable alternative for PE diagnosis in patients contraindicated for CTPA, supporting safe and accessible imaging strategies.

Coronary artery calcium (CAC) scoring plays a pivotal role in assessing the risk for cardiovascular disease events to guide the intensity of cardiovascular disease preventive efforts. Accurate CAC scoring from gated cardiac Computed Tomography (CT) relies on precise segmentation of calcification. However, the small size, irregular shape, and sparse distribution of calcification in 3D volumes present significant challenges for automated CAC assessment. Training reliable automatic segmentation models typically requires large-scale annotated datasets, yet the annotation process is resource-intensive, requiring highly trained specialists. To address this limitation, we propose S²CAC, a semi-supervised learning framework for CAC segmentation that achieves robust performance with minimal labeled data. First, we design a dual-path hybrid transformer architecture that jointly optimizes pixel-level segmentation and volume-level scoring through feature symbiosis, minimizing the information loss caused by down-sampling operations and enhancing the model's ability to preserve fine-grained calcification details. Second, we introduce a scoring-driven consistency mechanism that aligns pixel-level segmentation with volume-level CAC scores through differentiable score estimation, effectively leveraging unlabeled data. Third, we address the challenge of incorporating negative samples (cases without CAC) into training. Directly using these samples risks model collapse, as the sparse nature of CAC regions may lead the model to predict all-zero maps. To mitigate this, we design a dynamic weighted loss function that integrates negative samples into the training process while preserving the model's sensitivity to calcification. This approach effectively reduces over-segmentation and enhances overall model performance. We validate our framework on two public non-contrast gated CT datasets, achieving state-of-the-art performance over previous baseline methods. Additionally, the Agatston scores derived from our segmentation maps demonstrate strong concordance with manual annotations. These results highlight the potential of our approach to reduce dependence on annotated data while maintaining high accuracy in CAC scoring. Code and trained model weights are available at: https://github.com/JinkuiH/S2CAC.

This study evaluated the diagnostic accuracy and differential diagnostic capabilities of 12 Large Language Models (LLMs), one cardiac radiologist, and 3 general radiologists in cardiac radiology. The impact of the ChatGPT-4o assistance on radiologist performance was also investigated. We collected publicly available 80 "Cardiac Case of the Month" from the Society of Thoracic Radiology website. LLMs and Radiologist-III were provided with text-based information, whereas other radiologists visually assessed the cases with and without the ChatGPT-4o assistance. Diagnostic accuracy and differential diagnosis scores (DDx scores) were analyzed using the χ2, Kruskal-Wallis, Wilcoxon, McNemar, and Mann-Whitney U tests. The unassisted diagnostic accuracy of the cardiac radiologist was 72.5%, general radiologist-I was 53.8%, and general radiologist-II was 51.3%. With ChatGPT-4o, the accuracy improved to 78.8%, 70.0%, and 63.8%, respectively. The improvements for general radiologists-I and II were statistically significant (P≤0.006). All radiologists' DDx scores improved significantly with ChatGPT-4o assistance (P≤0.05). Remarkably, Radiologist-I's GPT-4o-assisted diagnostic accuracy and DDx score were not significantly different from the Cardiac Radiologist's unassisted performance (P>0.05).Among the LLMs, Claude 3 Opus and Claude 3.5 Sonnet had the highest accuracy (81.3%), followed by Claude 3 Sonnet (70.0%). Regarding the DDx score, Claude 3 Opus outperformed all models and radiologist-III (P<0.05). The accuracy of the general radiologist-III significantly improved from 48.8% to 63.8% with GPT4o assistance (P<0.001). ChatGPT-4o may enhance the diagnostic performance of general radiologists in cardiac imaging, suggesting its potential as a diagnostic support tool. Further studies are required to assess the clinical integration.

Breast magnetic resonance imaging is a critical tool for cancer detection and treatment planning, but its clinical utility is hindered by poor specificity, leading to high false-positive rates and unnecessary biopsies. This study introduces a transformer-based framework for automated classification of breast lesions in dynamic contrast-enhanced MRI, addressing the challenge of distinguishing benign from malignant findings. We implemented a SegFormer architecture that achieved an AUC of 0.92 for lesion-level classification, with 100% sensitivity and 67% specificity at the patient level - potentially eliminating one-third of unnecessary biopsies without missing malignancies. The model quantifies malignant pixel distribution via semantic segmentation, producing interpretable spatial predictions that support clinical decision-making. To establish reproducible benchmarks, we curated BreastDCEDL_AMBL by transforming The Cancer Imaging Archive's AMBL collection into a standardized deep learning dataset with 88 patients and 133 annotated lesions (89 benign, 44 malignant). This resource addresses a key infrastructure gap, as existing public datasets lack benign lesion annotations, limiting benign-malignant classification research. Training incorporated an expanded cohort of over 1,200 patients through integration with BreastDCEDL datasets, validating transfer learning approaches despite primary tumor-only annotations. Public release of the dataset, models, and evaluation protocols provides the first standardized benchmark for DCE-MRI lesion classification, enabling methodological advancement toward clinical deployment.

BackgroundRight ventricular (RV) function is a key factor in the diagnosis and prognosis of heart disease. However, current advanced CT-based assessments rely on semi-automated segmentation of the RV blood pool and manual delineation of the RV free and septal wall boundaries. Both of these steps are time-consuming and prone to inter- and intra-observer variability. MethodsWe developed and evaluated a fully automated pipeline consisting of two deep learning methods to automate volumetric and regional strain analysis of the RV from contrast-enhanced, ECG-gated cineCT images. The Right Heart Blood Segmenter (RHBS) is a 3D high resolution configuration of nnU-Net to define the endocardial boundary, while the Right Ventricular Wall Labeler (RVWL) is a 3D point cloud-based deep learning method to label the free and septal walls. We trained our models using a diverse cohort of patients with different RV phenotypes and tested in an independent cohort of patients with aortic stenosis undergoing TAVR. ResultsOur approach demonstrated high accuracy in both cross-validation and independent validation cohorts. RHBS and RVWL both yielded Dice scores of 0.96, and accurate volumetry metrics. RVWL achieved high Dice scores (>0.90) and high accuracy (>93%) for wall labeling. The combination of RHBS and RVWL provided accurate assessment of free and septal wall regional strain, with a median cosine similarity value of 0.97 in the independent cohort. ConclusionsA fully automated 3D cineCT-based RV regional strain analysis pipeline has the potential to significantly enhance the efficiency and reproducibility of RV function assessment, enabling the evaluation of large cohorts and multi-center studies. Key PointsO_LIRV endocardial segmentation of contrast enhanced CT scans can be utilized to perform volumetry, and when paired with labeling of free and septal walls, regional evaluation of surface strain. C_LIO_LIHowever, this has previously been performed using time-intensive semi-automated segmentation methods and manually labeling free wall and septal wall regions.. C_LIO_LIHere, we describe an automated, deep learning-based approach which uses two separate DL models to define the endocardial boundary (in 3D) and then label the free and septal walls on the endocardial surface. C_LIO_LIOur approach facilitates rapid and automatic advanced phenotyping of patients. This reduces prior limitations of potential interobserver variability and challenges associated with evaluating large cohorts. C_LI

To evaluate the translational capabilities of foundation models, we develop a pathological concept learning approach focused on kidney cancer. By leveraging TNM staging guidelines and pathology reports, we build comprehensive pathological concepts for kidney cancer. Then, we extract deep features from whole slide images using foundation models, construct pathological graphs to capture spatial correlations, and trained graph neural networks to identify these concepts. Finally, we demonstrate the effectiveness of this approach in kidney cancer survival analysis, highlighting its explainability and fairness in identifying low- and high-risk patients. The source code has been released by https://github.com/shangqigao/RadioPath.

Clinical machine learning (CML) in brain MRI analysis often assumes that "more data = better performance." However, when added samples derive from a different distribution than the training set, accuracy can decline--a phenomenon known as the Data Addition Dilemma. Here, we present the first systematic study of this dilemma in longitudinal traumatic brain injury (TBI) MRI, where acute baseline scans (session 1, S1) and follow-up scans (session 2, S2) exhibit pronounced distributional shifts. We make three key contributions. First, we quantify how intra-subject shifts (S1 [-&gt;] S2) and inter-subject variability jointly affect classifier performance in a 14-subject (28-scan) cohort spanning mild to severe TBI. Second, we compare four training schemes--(1) intra-session upper bound (S1 [-&gt;] S1), (2) cross-session OOD test (S1 [-&gt;] S2), (3) pooled training (S1+S2 [-&gt;] S1, S2), and (4) LOSO-IPA, which augments training with one unlabeled S2 scan per patient--using a lightweight logistic-regression model on five-component PCA features. Third, we derive actionable deployment insights: naive pooling can impair accuracy; pooled training trades baseline performance for robustness; and LOSO-IPA recovers near-intra-session accuracy. Accordingly, we recommend unlabeled per-subject follow-up anchoring and diagonal CORrelation ALignment (CORAL) covariance adjustment prior to inference. These findings clarify when additional data aid versus hinder CML in medical imaging and establish a minimally invasive framework for reliable longitudinal severity assessment in TBI.

Rheumatic heart disease (RHD) represents a significant global health challenge, disproportionately affecting over 40 million people in low- and middle-income countries. Early detection through color Doppler echocardiography is crucial for treating RHD, but it requires specialized physicians who are often scarce in resource-limited settings. To address this disparity, artificial intelligence (AI)-driven tools for RHD screening can provide scalable, autonomous solutions to improve access to critical healthcare services in underserved regions. This paper introduces RADAR (Rapid AI-Assisted Echocardiography Detection and Analysis of RHD), a novel and generalizable AI approach for end-to-end spatiotemporal analysis of color Doppler echocardiograms, aimed at detecting early RHD in resource-limited settings. RADAR identifies key imaging views and employs convolutional neural networks to analyze diagnostically relevant phases of the cardiac cycle. It also localizes essential anatomical regions and examines blood flow patterns. It then integrates all findings into a cohesive analytical framework. RADAR was trained and validated on 1,022 echocardiogram videos from 511 Ugandan children, acquired using standard portable ultrasound devices. An independent set of 318 cases, acquired using a handheld ultrasound device with diverse imaging characteristics, was also tested. On the validation set, RADAR outperformed existing methods, achieving an average accuracy of 0.92, sensitivity of 0.94, and specificity of 0.90. In independent testing, it maintained high, clinically acceptable performance, with an average accuracy of 0.79, sensitivity of 0.87, and specificity of 0.70. These results highlight RADAR's potential to improve RHD detection and promote health equity for vulnerable children by enhancing timely, accurate diagnoses in underserved regions.

To develop and evaluate the clinical feasibility of a dynamic computed tomography (CT) protocol for assessing patellofemoral (PF) and tibiofemoral (TF) kinematics before and after total knee arthroplasty (TKA), and to quantify postoperative kinematic changes in a pilot study. In this prospective single-centre study, patients with primary osteoarthritis scheduled for cemented TKA underwent dynamic CT scans preoperatively and at 1-year follow-up during active flexion-extension-flexion. Preoperatively, the femur, tibia and patella were segmented using a neural network. Postoperatively, computer-aided design (CAD) implant models were aligned to CT data to determine relative implant-bone orientation. Due to metal artefacts, preoperative patella meshes were manually aligned to postoperative scans by four raters, and averaged for analysis. Anatomical coordinate systems were applied to quantify patellar flexion, tilt, proximal tip rotation, mediolateral translation and femoral condyle anterior-posterior translation. Descriptive statistics were reported, and interoperator agreement for patellar registration was assessed using intraclass correlation coefficients (ICCs). Ten patients (mean age, 65 ± 8 years; 6 men) were analysed across a shared flexion range of 14°-55°. Postoperatively, the patella showed increased flexion (median difference: 0.9°-3.9°), medial proximal tip rotation (median difference: 1.5°-6.0°), lateral tilt (median difference: 2.7°-5.5°), and lateral shift (median difference: -1.5 to -2.8 mm). The medial and lateral femoral condyles translated 2-4 mm anterior-posteriorly during knee flexion. Interoperator agreement for patellar registration ranged from good to excellent across all parameters (ICC = 0.85-1.00). This pilot study demonstrates that dynamic CT enables in vivo assessment of PF and TF kinematics before and after TKA. The protocol quantified postoperative kinematic changes and demonstrated potential as research tool. Further automation is needed to investigate relationships between these kinematic patterns and patient outcomes in larger-scale studies. Level III.