Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Neuroimaging and cerebrospinal fluid analyses in humans reveal that loss of blood–brain barrier integrity and brain capillary pericyte damage are early biomarkers of cognitive impairment that occur independently of changes in amyloid-β and tau.
An algorithm trained on half a million electronic medical records predicts chronic kidney disease in diabetic patients using a small set of defined clinical features, outperforming predictions derived from clinical trial data.
A deep learning algorithm applied to the electrocardiogram—a test of the heart’s electrical activity—can detect abnormally low contractile function of the heart, opening up the possibility for a simple screening tool for this condition.
Analysis of electrocardiograms using an end-to-end deep learning approach can detect and classify cardiac arrhythmia with high accuracy, similar to that of cardiologists.
Expression of the exercise-induced myokine irisin (FNDC5) is lower in patients with AD. Whereas knockdown of FNDC5/irisin is sufficient to induce learning and memory deficits, restoration of its expression can ameliorate these phenotypes in rodent models.