The prevalence of vision disorders and eye cancer risk, driven by global population aging, necessitates precise ophthalmological disease screening, yet standardized and robust approaches for risk stratification remain absent. Here, we present SonoEye, a vision-language ultrasound foundation model for eye disease risk stratification, pre-trained on 215,356 image-text pairs from 70,452 patients using contrastive learning and fine-tuned with consensus-confirmed patient-level labels. The model achieves 98.3% screening sensitivity and a mean accuracy of 96.3% in differentiating 18 diseases. We established the Eye Reporting and Data System (Eye-RADS), a 4-tier risk stratification framework (normal, low-vision risk, high-vision risk, and tumor risk), demonstrating excellent Cohen's kappa agreement across internal (0.808) and external (0.677-0.685) test cohorts. Notably, incorporating age significantly improved Eye-RADS performance in elderly but not younger populations, indicating age-related features in eye care. Through image-text alignment, SonoEye generates structured clinical reports with interpretable attention-based visualizations, advancing automated vision risk stratification, particularly for aging populations in resource-limited settings.

Standardization of electronic health records (EHRs) has enabled the use of clinical codes in AI. We introduce ClinVec, an embedding store that provides embeddings for 153,166 clinical codes and concepts across eight vocabularies. ClinVec embeds ClinGraph, a knowledge graph with over 2 million edges tailored to clinical vocabularies used in EHRs. We validate the embeddings using an inter-institutional clinician panel and N = 3767 clinical term pairs spanning 11 disease areas, and we find that embedding similarity reflects clinical relatedness. We use ClinVec for knowledge injection in large language model medical question answering and for unsupervised patient stratification and risk prediction. By providing a shared representation of clinical concepts, ClinVec supports knowledge-grounded AI systems for modeling patients and populations.

Federated medical AI revolutionizes multi-center collaboration, while communication cost, data scarcity, and heterogeneity still limit its practical deployment. Foundation models (FMs) offer a promising avenue for addressing these challenges, owing to their generalization capabilities and efficient adaptability to medical tasks. Here, we present Federated Generative Prompt Learning (Fed-GPL), a universal and efficient framework for multi-center medical image analysis. It collaboratively trains a prompt generator that produces customized prompts for each patient, capturing patient-specific variations and enabling precise medical diagnosis. Fed-GPL is compatible with various vision FMs and medical tasks, such as Vision Transformer (ViT) for diabetic retinopathy and melanoma classification, and Segment Anything (SAM) for polyp and prostate segmentation. Fed-GPL outperforms traditional models and full fine-tuning methods, with only 8.26% and 6.55% of the total FM parameters being trained across classification and segmentation tasks, while converging within just 15 communication rounds. For low-resource settings, Fed-GPL maintains its performance with 5% of the original training data.

Sudden Cardiac Death (SCD) is a fatal event occurring within one hour of a witnessed or 24 hours of an unwitnessed Sudden Cardiac Arrest (SCA), being the leading medical cause of death among adolescent and young adult athletes. We examined the epidemiology of sports-related SCD (SrSCD) and SCA (SrSCA) incidence in adolescents and young adults, explainable Artificial Intelligence (xAI) applied to life-threatening arrhythmias, and cardiac electrophysiological models. We systematically searched peer-reviewed studies from eight databases between 2013-2025 (PROSPERO: CRD42024565960), using PROBAST for bias assessment. From 9574 studies, we included 84 (incidence: 16, xAI: 30, modelling: 38). SrSCD incidence ranged from 0.1 to 0.6 per 100,000 participants per year. Gradient-weighted Class Activation Mapping dominated as xAI technique. Cardiac electrophysiological models predominantly focused on cellular and tissue-level electrophysiology. We advocate for standardised SrSCD/SrSCA definitions and integration of epidemiological risk factors with xAI and cardiac modelling frameworks to advance athlete-specific risk stratification.

Physical frailty and concern about falling are interrelated geriatric conditions that significantly impair mobility, limit daily activity, and reduce life quality among older adults. Traditional methods for assessing frailty and fall concern rely heavily on questionnaires such as the Fried Frailty Phenotype (FFP) and the Falls Efficacy Scale-International (FES-I). These tools, although clinically established, are subjective, require professional administration, and cannot capture the mutual influence between the two conditions. In this study, we propose a wearable-sensor-based, objective framework that simultaneously predicts physical frailty status and level of concern about falling by analyzing real-world activity patterns. We collected continuous chest acceleration data from 146 participants over a 48-hour period using a pendant sensor. These activity sequences were transformed into barcodes and analyzed using global complexity metrics (e.g., single- and multi-scale entropy) and local sequential dynamics via bidirectional LSTM networks. A multi-task deep learning model with attention mechanisms was proposed to jointly predict frailty and fall concern. Our model achieved high predictive performance for both tasks, achieving F1 scores of 93.12% for physical frailty and 86.27% for concern about falling. This study provides the first joint modeling of physical frailty and concern about falling using long-term real-world accelerometry data.

OBJECTIVES: To evaluate the diagnostic accuracy of magnetic resonance imaging (MRI) compared to histopathology for detecting extramural venous invasion (EMVI) in rectal cancer, and to assess their prognostic value for long-term recurrence-free survival (RFS) and overall survival (OS). MATERIALS AND METHODS: In a prospective cohort of 150 patients with biopsy-confirmed rectal adenocarcinoma, EMVI was assessed via high-resolution 1.5-T MRI and histopathological examination. Interobserver agreement, clinical correlations, and survival outcomes were analyzed. Kaplan-Meier and Cox regression analyses were used to evaluate recurrence-free survival (RFS) and overall survival (OS). RESULTS: mrEMVI was detected in 52.7% of patients and showed good interobserver agreement (κ = 0.68). pEMVI was confirmed in 45.9% of surgical cases, with fair agreement with mrEMVI (κ = 0.40). mrEMVI was significantly associated with advanced tumor stage, larger tumor volume, synchronous metastases, and elevated CEA levels. mrEMVI positivity predicted poorer RFS (HR 5.22, p < 0.001) and OS (HR 3.40, p = 0.01), while pEMVI showed no significant prognostic value. CONCLUSION: mrEMVI is a more reliable imaging biomarker strongly correlated with long-term oncologic events compared to histopathological EMVI. KEY POINTS: Question Reliable preoperative markers are lacking for accurate EMVI detection in rectal cancer, creating uncertainty in risk stratification and treatment decisions due to inconsistent MRI-pathology concordance. Findings mrEMVI showed strong prognostic value for recurrence and survival, whereas pEMVI did not, and MRI offered higher reliability and reproducibility than histopathology. Clinical relevance Identifying mrEMVI preoperatively enables more accurate risk stratification, helping clinicians tailor treatment intensity and potentially improve survival, while highlighting MRI as a more reliable tool than pathology for guiding management decisions in rectal cancer.