AFP Assistant is a multilingual chatbot designed to improve acute flaccid paralysis and polio surveillance in low-resource Ethiopian language communities.

HemaGuide shows how large language models can become more clinically useful when grounded in guidelines, tumor board memory, literature retrieval, and molecular interpretation tools.

Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from NPJ digital medicine. MAIN Today we’re looking at “FedFound: a federated foundation model for lifespan brain…

A JAMA Network Open study suggests that large language models can help screen emergency department eTrigger cases for missed opportunities for diagnosis, but they are not ready to…

A multicenter retrospective study reports a machine learning-based calculator that predicts initial severity and time-specific mortality in adult secondary hemophagocytic lymphohistiocytosis.

Recent Orthopedics AI coverage is moving beyond model accuracy toward clinical fit: routine-data screening, prognosis across settings, engineered implants and delivery systems, and harder questions about trust.

Recent respiratory AI coverage points toward deployment pressure: imaging at the bedside, calibrated asthma risk, genomic TB resistance, and workflow-aware models built for decisions, not just prediction.

Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from Nature medicine. MAIN Today we’re looking at a pragmatic, cluster-randomized trial titled “Generative AI-enabled clinical decision…

A new ultrasound-specific instruction dataset and benchmark show that domain-aligned training can substantially improve compact vision-language models for ultrasound understanding.

Baker, Frommer, Uslu et al. report an AI-driven strategy for CAR T target discovery and identify GPNMB as a candidate antigen spanning multiple cancers.