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.
Graph neural networks offer a unifying artificial intelligence framework to model related objects, ranging from tissue architecture and geometrical relationships to patient similarity and multi-organ networks. Applications of this technology in nephrology include computational representation of kidney histopathology and modelling of the complex interactions between organs in kidney diseases.
Deep Learning (DL) holds great promise to improve patient outcomes by improving the precision and speed of disease diagnosis and treatment recommendations. Given the efficacy of DL in image analysis, pathology will likely be one of the first medical fields transformed by DL. However, several challenges must be overcome before we can expect to see the use of DL transform the digital future of pathology.
The increasing prevalence of chronic kidney disease (CKD) is placing a growing burden on healthcare systems, which results in considerable economic and environmental challenges. Sustainable CKD care and optimization of patient outcomes requires a new approach to the organization of healthcare systems, in which home monitoring will have a pivotal role.
