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.
An image reconstruction method leveraging the capture of 511 keV annihilation photons and prompt γ-ray emission in the same energy window allows for the simultaneous in vivo imaging of two radiotracers for positron emission tomography.
The development of machine-learning systems for safer, robust and fairer outcomes should leverage fine-tuning, generalization, explainability and metrics of uncertainty.
Functionally distinct T-cell populations can be generated from T cells that received the same stimulation by altering the viscoelasticity of their surrounding extracellular matrix.
Posterior scleral birefringence, measured by triple-input polarization-sensitive optical coherence, predicts the onset of myopia in guinea pigs and is associated with myopia status in patients.
Fusing the anti-inflammatory enzyme indoleamine 2,3-dioxygenase to the tissue-anchoring protein galectin-3 ameliorates inflammation at the injection site while avoiding systemic immune suppression, as shown in multiple rodent models of inflammation.
A wearable electrochemical patch for the real-time monitoring of the biomarker C-reactive protein in sweat detects elevated concentrations of the protein in patients with acute or chronic inflammation.
An automated plaque assay leveraging lens-free holographic imaging and deep learning rapidly and accurately detects the cell-lysing events caused by viral replication.
A transformer-based representation-learning model that processes multimodal input in a unified manner outperformed non-unified multimodal models in two clinical diagnostic tasks.
Micropillar patterns causing changes in the nuclear and cellular morphologies of human mesenchymal stromal cells influence the conformation of the cells’ chromatin and their osteogenic differentiation in vitro and in mice.
A nanoparticle integrating a fusion protein of apolipoprotein A1 and interleukin-4 and that targets myeloid-cell-rich haematopoietic organs resolves immunoparalysis in ex vivo and in vivo models of sepsis.
T cells with a chimaeric antigen receptor specific for fluorescein isothiocyanate can be directed against solid tumours via the intratumoural administration of a fluorescein isothiocyanate-conjugated amphiphile that inserts itself in cell membranes.
A representation-learning strategy for machine-learning models applied to medical-imaging tasks improves model robustness and training efficiency and mitigates suboptimal out-of-distribution performance.
Demultiplexing PET–MRI data of solid tumours using machine learning allows the spatial characterization of intratumour tissue heterogeneity in mice and humans. Predicted maps of tissue subtypes within the tumour could aid in conducting image-guided biopsies and provide valuable insights linking the outcome of cancer therapies with phenotypic heterogeneity.
Intratumoural heterogeneity can be characterized spatially in patients with liver metastases from colorectal cancer via phenotype-specific multi-view learning models trained with PET–MRI data from mice with subcutaneous colon tumours.
The algorithmic detection of cancer-associated variants can be accelerated by leveraging machine-learning classifiers to filter out reads matched to pan-genome k-mer sets.
Amphiphilic peptides can aid the delivery of CRISPR ribonucleoproteins into primary human lymphocytes at low toxicity, boosting editing yields with respect to the use of electroporation.