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 in combination with human expertise could be the optimal approach to improving diagnostic accuracy while maintaining a safety net in clinical imaging.
Adoptive transfer of convalescent donor-derived SARS-CoV-2-specific T cells was safe and conferred faster recovery in patients with severe COVID-19 compared to standard of care.
This review explains how large language models (LLMs), such as ChatGPT, are developed and discusses their strengths and limitations in the context of potential clinical applications.
A collaboration system helps to integrate decisions between human experts and AI to optimize screening and triaging and to reduce clinicians’ workload.
CSF MTBR-tau243 is more related to tau tangles and clinical cognitive impairment in Alzheimer’s disease than phospho-tau biomarkers, which are more related to amyloid plaques.
Genome-wide polygenic scores quantify inherited risk by integrating information from many common DNA variants and hold considerable promise for enabling personalized medicine. By integrating information on coronary artery disease (CAD) and CAD-related risk traits from genetic datasets that were larger and more diverse than those used in the past, we developed an improved multi-ancestry polygenic predictor for CAD.
This ecological analysis using the Eurostat database estimated that summer 2022, the hottest summer on record, was associated with over 61,000 heat-related deaths across 35 countries in Europe, with the highest mortality rates in countries near the Mediterranean Sea.
Prospective assessment of a T-cell-inflamed gene signature and tumor mutational burden is feasible in patients with non-small cell lung cancer randomized to receive various anti-PD-1-based treatment combinations, with an encouraging objective response rate for first-line pembrolizumab + lenvatinib in one biomarker-defined group.
Strategies to monitor and modulate gut microorganisms to improve immunity and responses to cancer immunotherapy are increasingly available, with opportunities to expand and iterate on these approaches to treat, intercept and prevent cancer (and other diseases) altogether.
Serological analysis and infection outcomes of participants in the multi-center, prospectively enrolled OCTAVE cohort, comprising 2,686 participants with immune-suppressive diseases who recieved two COVID-19 vaccines, reveals specific clinical phenotypes that might benefit from specific COVID-19 therapeutic strategies.
A polygenic risk score for coronary artery disease developed using data from individuals of five different ancestries has increased accuracy across diverse populations.
In children with relapsed or refractory B cell acute lymphoblastic leukemia and in complete remission after CD19 CAR-T cell therapy, long-lived CAR-T cells express a persistence gene signature that is also present in persistent CD19 CAR-T cells from adults with chronic lymphocytic leukemia.
In patients with advanced melanoma, fecal microbiota transplantation from healthy donors combined with the anti-PD-1 inhibitors nivolumab or pembrolizumab was well tolerated with an encouraging objective response rate of 65% in the first-line treatment setting.
A risk score developed using biological, psychological and social factor data from the UK Biobank can predict different pain conditions, the risk of chronic pain spreading across body sites and the prognosis of chronic pain up to 9 years later.
