Principal Investigator: Dr Keith Smith
Institution: University of EdinburghTags: 41492, brain connectivity, Machine Learning, Physiological networks
We will conduct a three-year study on the patterns of integrated biological information contained in the UK BioBank and how these patterns are related to clinical outcomes of disease. This will involve advanced mathematical methods.
We will first integrate information of the structure of the brain- the connections between brain regions made through physical fibres- with the function of the brain- the interactions found between function related activations of these brain regions- to gain a deeper understanding of how the brain works. Novel methods will be applied to assess the patterns within these connections as well as how these patterns change over time. This will aid in understanding the deterioration of brain structure and function in diseases such as dementia as well as provide tools which will potentially aid in the detection and even prediction of these diseases.
Zooming out, we will then construct networks of connections between BioBank participants based on biological information obtained throughout the body as well as on information of diet and lifestyle. This will construct an atlas of human physiology which will be the basis for informing on links between health and disease using information on clinical outcomes. We expect, for instance, that patients who go on to develop dementia will take up similar positions within the BioBank network and, it follows, that participants without any clinical outcome information taking up a similar position may have undiagnosed dementia.
Due to the large scale of the UK BioBank, the collective information obtained from this approach can help uncover opaque patterns- unobvious even to a trained eye- in a highly robust manner. We hope that this will help to inform on the detection and prediction of diseases beyond what is possible in traditional health studies.