An artificial intelligence model applied to standard 12-lead ECGs detected a dilated cardiomyopathy phenotype and predicted its future onset across multiple cohorts. Its performance in relatives undergoing familial…
An electrode-free ring that senses skin activity and other physiological signals classified experimental states in a small study. Its local AI processing also reduced energy use compared with…
Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from Nature communications. MAIN Let’s start with a familiar ICU problem. A patient is on a ventilator,…
In a real-world digital obesity intervention, weight at four weeks was the strongest source of information for predicting later weight loss. Early app engagement, by contrast, was more…
Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from The lancet. Diabetes & endocrinology. MAIN When an adult is diagnosed with type 1 diabetes, the…
AnySleep combines flexible EEG and EOG inputs with high-resolution predictions to stage sleep across heterogeneous datasets. In a large retrospective study, it modestly improved conventional staging over a…
Global change may make the environment less favorable for terrestrial mammals across much of the world, while creating more favorable conditions elsewhere. An explainable AI analysis maps these…
A correction to a longitudinal study of AI-powered clinical documentation updates the characteristics of participating clinicians, but it does not report whether DAX Copilot improved efficiency. The revised…
An artificial intelligence method applied to noncontrast cardiac CT improved cardiovascular mortality prediction beyond coronary artery calcium scoring. The added information was particularly relevant in participants with no…
A deep-learning algorithm detected roughly seven in ten eyes with perimetric glaucoma in an external population-based validation. Performance was promising, but missed cases tended to have milder visual…