Associated organisationsKing's College London · Radboud Universiteit Nijmegen · University of OxfordEurope PMC affiliations are not treated as award recipients or mapped locations.
Funding£4.1M
PeriodJun 2019 — May 2024
In plain English
AI plain-English summary
Brain scans from hundreds of thousands of people are piling up in databases like UK Biobank, but researchers lack the tools to fully mine them. Current analysis methods cannot handle the sheer volume, the multiple imaging types, or the linked health and genetic data these studies produce. This project aims to build the computational platforms and statistical models needed to extract meaningful information from these massive datasets. The researchers will develop integrated analysis techniques that combine measures of brain structure, connectivity, and function, then apply machine learning to identify biologically interpretable markers of neurological disease. If successful, the work could transform how we understand variation in brain health across populations and improve the ability to characterise individual patients. For clinicians, this might mean more precise diagnostic tools for conditions like dementia or multiple sclerosis. For the broader public, the impact would be invisible but significant: better algorithms running behind the scenes in medical imaging software, enabling earlier detection and more personalised treatment planning without requiring new hardware or invasive procedures.
View original technical description
Neuroimaging enables the mapping of many aspects of the brain’s anatomy, connections, and function. New landmark studies including UK Biobank and the Human Connectome Projects are taking neuroimaging to the scale of populations. Such studies deploy multiple imaging modalities with the aim of learning more about the brain, and identifying imaging markers relevant to neurological disease. However, we cannot currently take full advantage of the richness of these new resources, including: major advances in the quality of data; complementarity of multi-modal imaging; large subject numbers; and linked information about health outcomes, genetics and risk factors. Our vision is to extend the reach of imaging neuroscience. This requires new research across multiple domains: from integrated cross-modal analysis, to more detailed and biologically-interpretable markers, to machine learning. We aim to enable neuroimaging to achieve its full potential, from the modelling of variation in populations to the characterisation of individual subjects. We will deliver powerful modelling approaches and research platforms, continuing our long track record of disseminating software for use by basic and clinical neuroscientists. Further, we will leverage our leadership in big data projects to demonstrate how these approaches and computational tools can advance our understanding of the brain and its diseases.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know