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.
Leveraging or improving established technology for clinical imaging to extract additional physiological information can enhance the quality of the subsequent assessments and widen the technology’s uses.
The development of machine-learning systems for safer, robust and fairer outcomes should leverage fine-tuning, generalization, explainability and metrics of uncertainty.
Graph neural networks and transformers taking advantage of contextual information and large unannotated multimodal datasets are redefining what is possible in computational medicine.
Research manuscripts and the associated scientific data generated for projects that are funded by federal agencies in the United States will need to be made publicly available immediately on publication.
By allowing for the visualization of living tissue faster, at higher contrast or with larger fields of view, imaging modalities widely used in research are making inroads into the detection of disease in the human body.
Better cell sourcing and increasingly fine control over cell differentiation, tissue formation and cell and tissue maturation are pushing forward progress in disease modelling, drug development and regenerative medicine.
Lessons being learned about the utility of COVID-19 diagnostics are informing the design, required real-world performance and deployment needs of technologies for the detection of infectious diseases.
The growth of citations to published content typically follows an S-shaped curve. We look back at the fairly homogeneous citation-growth patterns — and at the few exceptions to them — for the content that we published in 2017.