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.
Clinical implementations of machine learning that are accurate, robust and interpretable will eventually gain the trust of healthcare providers and patients.
Modelling diseases of the central and peripheral nervous systems and effectively treating neurological disorders via neuronal manipulation requires far better biomaterials and technology than are currently available.
For cell therapies to transition from promises to products, increased efforts need to be put into the identification of the factors and biological mechanisms that affect safety and efficacy, and into the design of cost-effective methods for the harvesting, expansion, manipulation and purification of the cells.
When designing translationally relevant delivery strategies to overcome the physicochemical and biological barriers to getting therapeutics into the right tissues and cells, building on tried-and-tested concepts often pays off.
Accurate diagnostics need technology — from imaging hardware and image reconstruction to machine learning — to detect markers associated with the cause of disease.
The translational achievements, commercialization prospects and eventual clinical and societal impacts of biomedical work accrue over a long time. They should be better captured and publicized.