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.
Artificial intelligence (AI) approaches can be employed to classify medical imaging data, with the hope that these systems might aid in clinical diagnoses. In this issue, Eric Oermann and colleagues apply deep learning to triage and prioritize radiological workflows to accelerate time to diagnosis for urgent neurological events. In a separate study, Olaf Ronneberger and colleagues report on a deep-learning framework that can recommend how patients should be referred for treatment for over 50 eye diseases with the same accuracy as expert ophthalmologists. Conor Liston discusses the implications of these studies in an accompanying News & Views. The cover depicts abstract 'layers' through the human eye, making the connection between layers of the retina and layers within a neural network and demonstrating the transparency of the approach developed by Ronneberger and colleagues.
To facilitate access to and improve the discoverability of the data in our papers, Nature Medicine is making the data availability statement in our papers more prominent and its language more transparent.
To address a critical roadblock that can occur in translational and clinical research, the National Cancer Institute and the Food and Drug Administration, in coordination with the DREAM Challenges, are launching the first computational challenge using multi-omics datasets to detect and correct specimen mislabeling.
Machine learning can be used for computer-aided diagnosis of acute neurological events and retinal disease and can be incorporated into conventional clinical workflows to improve health outcomes.
Impaired de novo NAD+ biosynthesis predisposes to acute kidney injury, and augmenting NAD+ metabolism with oral nicotinamide supplementation may prevent acute kidney injury.
Preconception cold–induced alterations of sperm DNA methylation result in offspring with altered brown adipose tissue and improved adaptation to overnutrition and hypothermia.
Enhancer profiling of breast tumors reveals that chromatin regulatory elements contribute to the clonal fitness landscape, treatment resistance and phenotypic divergence.
An in vivo cellular reprogramming strategy to generate epithelial cells from wound mesenchymal cells promotes healing and provides a new avenue for the treatment of nonhealing wounds.
Antibiotic treatment of SIV-infected nonhuman primates shows that inducing bacterial dysbiosis in the gut—similar to that seen in HIV infection—does not promote disease progression, questioning its potential role in progression to HIV/AIDS.
In utero GBA gene therapy extends lifespan and provides long-lasting phenotypic amelioration in a mouse model of neuronopathic Gaucher disease. Fetal ultrasound-guided in utero gene vector delivery is also achieved in the non-human primate brain.
A glutamate-gated chloride channel delivered via gene therapy is shown to detect elevated brain glutamate levels and trigger the suppression of neuronal excitability, thereby attenuating seizure activity in two rodent models of epilepsy.
A deep-learning algorithm is developed to provide rapid and accurate diagnosis of clinical 3D head CT-scan images to triage and prioritize urgent neurological events, thus potentially accelerating time to diagnosis and care in clinical settings.
A novel deep learning architecture performs device-independent tissue segmentation of clinical 3D retinal images followed by separate diagnostic classification that meets or exceeds human expert clinical diagnoses of retinal disease.
The muscle-secreted, exercise-induced peptide hormone apelin decreases with aging and sarcopenia, and its repletion in aged mice with recombinant protein improves muscle mass and function.
Using 13C-labeled substrates in vivo, this group shows that metformin inhibits mG3PDH to reduce hepatic gluconeogenesis and lower glycemia by altering the redox potential of the cytosol of hepatocytes rather than affecting substrate availability.
The antidiabetic action of metformin raises heptocyte intracellular AMP levels, causing allosteric inhibition of fructose-1,6-bisphosphatase and thus reductions in gluconeogenesis.
Mechanism-based small-molecule inhibitors targeting a gut microbial enzyme lower circulating levels of the prothrombotic metabolite trimethylamine-N-oxide and suppress diet-induced thrombosis in mice.
Interleukin-1β promotes an atheroprotective phenotype in late-stage lesions of mice, suggesting the possibility of deleterious effects of interleukin-1β blockade in the setting of myocardial infarction.
Early and prolonged administration of antiretroviral therapy to SIV-infected and post-exposure-vaccinated rhesus macaques was associated with absence or delay of detectable virus after therapy interruption.
A blood-based DNA sequencing assay to infer tumor mutational burden in the absence of tumor biopsy predicts response to PD-L1 blockade in patients with non-small-cell lung cancer.
Microsatellite instability and Epstein–Barr virus positivity represent novel biomarkers of clinical response to PD-1 blockade in patients with metastatic gastric cancer.
Patients with glioblastoma experience lymphopenia and sequestration of T cells in the bone marrow, which is recapitulated in mice with brain tumors, where the reversible nature of this effect is demonstrated by an approach that enables the efficacy of other immunotherapeutics.
Topographic analysis of the active regulatory landscape in estrogen receptor-positive breast cancer uncovers a role for transcription factor YY1 in modulating phenotypic heterogeneity during tumor evolution and endocrine resistance.