A portable, non-ionizing radio-frequency brain scanner paired with deep learning distinguished hemorrhagic and ischemic stroke from non-target presentations in an interim multicentre clinical evaluation. In prospectively held-out scans,…

REM sleep behavior disorder (RBD) can coexist with major depressive disorder (MDD), yet confirmation requires video-polysomnography (v-PSG). In this cross-sectional study, a multimodal AI model using brief smartphone-recorded…

Routine H&E whole-slide images may contain clinically useful signals of immune inflammation in clear-cell renal cell carcinoma (ccRCC). In this cohort study, a deep-learning CD8+ T-cell signature identified…

Recent Cardiology coverage shows AI moving beyond isolated detection: ECG, imaging, wearables, and biological signals are being tested as scalable tools for earlier risk phenotyping, with generalizability now…

In three Chinese cohorts, sex-specific transformer-based models predicted 10-year cardiovascular risk more accurately than matched Cox models and established risk scores. China-AIHeart’s external validation suggests that population-specific, time-to-event…

FlowXAI is an interpretable decision-support system for classifying B-cell non-Hodgkin lymphoma (B-NHL) from multiparameter flow-cytometry data. In a large computational evaluation, its diagnostic confidence categories tracked with observed…

PRISM is a large MRI-specific foundation model designed to separate anatomy from sequence-related contrast. Across 44 downstream tasks, it ranked first in 39, indicating that broad multi-sequence pretraining…

Target trial emulation improved the structure of observational analyses, but it did not recover randomized-trial mortality effects for two widely studied heart-failure treatments. The findings underscore how strongly…

This week’s strongest papers share a hard lesson: medical AI earns its place not by seeing more data, but by making reliable decisions amid missing inputs, noisy signals,…

Recent Nephrology & Urology coverage shows the field’s center of gravity shifting: multimodal performance matters, but portability, alert burden, workflow fit, and decision consequences now determine whether AI…