Generative AI can transform intraoperative frozen sections of skin cancer margins into convincing FFPE-like images while preserving the speed of frozen-section assessment. In a retrospective study, this approach…

When an electronic deterioration score directly triggered rapid response team notifications, risk-adjusted inhospital mortality fell among high-risk adult inpatients. The study suggests that the clinical workflow surrounding an…

Recent Pediatrics coverage shows AI moving closer to real clinical decisions, while youth-facing tools expose a harder frontier: calibration, context, trust, privacy, and proof that promising outputs improve…

GPT-5 can convincingly sound like different oncology specialists without producing genuinely distinct specialist decisions. In a simulated gastrointestinal tumour board, persona prompting changed linguistic style far more than…

Neurological AI is moving toward accessible monitoring, individualized prognosis, and learned biological signatures. Yet this coverage points to a sharper constraint: dependable decisions require better measurements and stronger…

A routine chest radiograph may contain enough information to help identify patients at risk of developing chronic kidney disease (CKD). In a large two-centre US cohort, CXR-CKD5 predicted…

A generative AI model may produce realistic-looking forecasts of pediatric brain tumor growth without yet providing reliable evidence for treatment planning. Li, Sun, and Pei identify two central…

A modular artificial-intelligence system accurately identified the pollen or fungal allergens associated with seasonal allergic rhinitis and reproduced expert allergen immunotherapy decisions. Its performance remained high when models…

Artificial intelligence applied to routine electrocardiograms may help identify which older adults are most likely to benefit from atrial fibrillation screening. In a post hoc analysis of the…

Recent Radiology & Imaging coverage points beyond isolated detection models: toward reusable foundations, workflow-aware assistance, quantitative biomarkers, and systems designed to remain reliable after deployment.