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.
A prognostic model for invasive breast cancer that is based on interpretable measurements of epithelial, stromal, and immune components outperforms histologic grading by expert pathologists. This model could improve clinical management of patients diagnosed with invasive breast cancer and address the concerns of pathologists about artificial intelligence (AI) trustworthiness by providing transparent and explainable predictions.
Nature Medicine asks six leading AI researchers to explain how LLM-powered chatbots are having an impact on health, from virtual nurses to detecting cancer progression.
In people with cardiovascular disease (without diabetes) and overweight or obesity, the GLP-1 receptor agonist semaglutide significantly reduced the risk of major cardiovascular events.
A study of nearly one million people who underwent a CT scan before 22 years of age finds that the radiation from CT scans increased the risk of hematological malignancies in a dose-dependent manner. These findings highlight the continued need to justify CT scans and minimize radiation doses.
The adoption of medical AI devices in clinical practice is still in its infancy, with uneven usage across procedures and locations in the USA, which underscores the inequities in its real-world implementation.
The prevalence of aneuploid cells in miscarried human embryos is higher than previously quoted. Genomic imbalances seem to be less tolerated in the embryoblast than the trophoblast, which indicates that allocation of aneuploid cells to the inner cell mass during blastocyst formation might have a detrimental effect on embryo development.
Deep learning enables comprehensive and interpretable scoring for breast cancer prognosis prediction, outperforming pathologists in multicenter validation and providing insight on prognostic biomarkers.
A four-arm cluster-randomized trial testing the effects of a low cash incentive, a high cash incentive, a health message and a placebo on COVID-19 vaccine uptake in villages in rural Ghana demonstrated highest uptake with a low cash incentive.
Preclinical and initial human studies suggest that glucagon-like peptide-1 receptor agonists may be promising treatments for alcohol use disorder, but existing approved treatments should be used until safety and efficacy is demonstrated in clinical trials.
We find that people with non-suppressible human immunodeficiency virus (HIV) viremia despite antiretroviral therapy (ART) exhibit several distinguishing features. These include expanded CD4+ T cell clones containing HIV proviruses integrated into transcriptionally permissive regions, the presence of certain proviral defects or human leukocyte antigen (HLA)-escape mutations, enhanced survival signatures, and muted interferon and cytotoxic CD8+ T cell responses.
Genomic analysis of products of conception collected after spontaneous pregnancy loss in the first trimester reveals previously undetected chromosomal aberrations and a higher degree of mosaic chromosomal imbalances.
New data show that AI could enhance imaging-based screening for pancreatic cancer; however, its evaluation must be rigorous and adhere to the same standards used for conventional screening.
In a phase 2 trial, the oral type II RAF inhibitor tovorafenib exhibited an overall response rate of 67% in patients with BRAF-altered relapsed/refractory pediatric low-grade glioma.
