This study shows that automated meningioma segmentation can be paired with clinically interpretable uncertainty estimates, not just point predictions.

A multimodal AI system integrating ultrasound, CT, MRI, clinical variables, and tumor markers achieved performance close to multidisciplinary team consensus for preoperative ovarian cancer assessment.

A deep learning model trained on routine breast cancer histology predicted germline ATM pathogenic variant carrier status, suggesting that morphology may carry inherited-risk signals.

Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from Alzheimer's & dementia : the journal of the Alzheimer's Association. MAIN The paper is titled “Digital…

UNICORN is a transformer-based model designed to integrate multiple histological stains from coronary artery whole-slide images for atherosclerosis severity classification.

A large mixed-methods study of dementia research finds persistent gender disparities in publication output, citation impact, and senior authorship.

A 24-hour single-lead Holter ECG may contain enough signal to identify patients at elevated five-year risk of heart failure. In this cohort study, DeepHHF used day-long ECG modeling…

FetalCLIP is a fetal ultrasound-specific vision-language foundation model designed to learn reusable image representations from paired ultrasound images and text.

PEAL offers a one-shot federated learning approach that reproduces pooled individual-level linear mixed-model results while keeping patient records local.

ARTIMES proposes a more reproducible way to measure pleural mesothelioma response on CT, replacing difficult linear measurements with AI-assisted tumour volume assessment.