Several members of the lab are involved in a long-term research collaboration between the University of Oxford Big Data Institute and the pharmaceutical company Novartis. The aim of the collaboration is to develop statistical machine learning methods to better understand complex diseases using Novartis' clinical trial data. The partnership is focusing on flagship programs in multiple sclerosis (MS), rheumatology and dermatology. MS is a chronic and ultimately debilitating disease that affects approximately 2.5 million individuals worldwide. A disease of the central nervous system, MS is characterised by the inflammation and eventual destruction of the axons. Novartis has amassed a vast database from clinical trials targeting MS, including brain MRI images across multiple modalities, clinical and genomic data. Using this data, the project aims to better characterise MS over the span of the disease and find biomarkers for early diagnosis, monitoring and prognosis of individual MS patients. The rheumatology and dermatology program will focus on the following autoimmune disorders: ankylosing spondylitis, rheumatoid arthritis, psoriatic arthritis and psoriasis. The goal of the study is to analyse the relevant studies and identify new factors driving disease progression and therapeutic response and develop cutting-edge tools to support clinical decision making.

Chris is co-investigator on two EPSRC programme grants: Bayesian Data Science for Health Research (Bayes4Health) and Computational Statistical Inference for Engineering and Security (CoSInES). The Bayes4Health project aims to develop new approaches to Bayesian data science that are needed both within the health sciences and more widely. By building on recent breakthroughs in efficient Monte Carlo integration methods for large data, and on new paradigms for Bayesian-like updates that are suitable for complex models, the programme will address key research challenges in the health sciences - directly developing new insights and understanding for these. The CoSInES project brings together a world class team of researchers from Universities of Warwick, Bristol, Lancaster, Oxford and the Alan Turing Institute to tackle fundamental challenges in Computational Statistics and key applications from Engineering and Security. The problems that CoSInES tackle are characterised by complex large data sets indexed in space and often continuously in time. To ensure their utility, the approaches developed are being accompanied with new theory on computational and statistical robustness.

The lab is involved with a long-term research collaboration between the pharmaceutical company Novo Nordisk and the Universities of Oxford and Manchester. The purpose of the project is to explore and develop novel statistical methods within AI and machine learning to support the increased use of data to generate knowledge and insights. The specific focus will be to advance the techniques in the area of multimorbidity. Patients with multimorbidity have complex medical needs and often experience a lower quality of life, higher hospital use, risk of developing more long-term conditions (e.g. depression), and death. This drives health care costs and creates challenges in coordinating health services. Moreover, much of the traditional evidence base on therapeutics arise from studies targeting single conditions. The aim of the collaboration is to explore the application and development of data science and statistical machine learning methods that provide insight, understanding and characterisation of longitudinal multimorbidity in the context of metabolic disease. The methods will exploit data from UK BioBank (UKBB) as well as large primary care data records (such as CPRD).

Several lab members are also part of the Alan Turing Institute - Royal Statistical Society statistical modelling and machine learning laboratory (Turing-RSS Lab), headed up by Chris. The purpose of the lab is to partner with the Joint Biosecurity Centre (JBC) to provide further statistical modelling and machine learning expertise in support of the government’s response to COVID-19, and to co-ordinate and strengthen modelling work across the Department of Health and Social Care (DHSC) Test and Trace programme. The Turing-RSS Lab provides independent insight and analysis of NHS Test and Trace data to grant the JBC deeper understanding of how the virus is spreading across the country and the epidemiological consequences. Statistical modelling helps data scientists to predict what the virus might do next, based on what is understood about it already.