Principal Investigator: Professor Stefan Neubauer
Institution: University of Oxford
Professor Steffen Petersen, Queen Mary University of London, London, UK
Dr Matt Kelly, Perspectum Diagnostics Ltd, Oxford, UKTags: 59867, Brain, cardiovascular, Liver, multimorbidity, multivariate statistical machine learning, risk prediction
Diseases of the heart, arteries and brain are leading causes of illness and death. Furthermore, 7 out of 10 people with liver disease are not aware that they have a liver condition until their first hospital admission to accident and emergency.
Current statistical methods are too simple, for example, assuming that different organ systems do not interact with each other. In this work we will make use of a very large set of measurements of health and improved statistical methods, to combine data from various organ systems together with general information (such as age, gender etc) and environmental factors. We will explore all possible ways in which these different measurements relate to each other. Current tools that aim to predict the risk of developing diseases are based on only a few measurements of single organs or simple blood tests, and do not work well. So as a second goal, we will test if a more inclusive approach, taking into account all the factors identified by our exploratory work, can lead to improved predictions of disease development. We aim to develop novel and robust combined risk assessment tools that can predict medical conditions more accurate and timely. Our work should assist medical practitioners and society in general by improving health care decision making. Ultimately, we hope to support the UK Biobank’s aim to improve the prevention and diagnosis of diseases worldwide.