Three MRC-funded UKRI Innovation/Rutherford Fellowships awarded at IHI, as part of the new HDR-UK Institute

Dr Ruth Blackburn. During the three-year fellowship Ruth will examine the distribution and dispersion of health behaviours in social groups, at scale, using a range of electronic health records and education data. Self-harm behaviours associated with high-volume healthcare usage (intentional self-injury, substance misuse and violence) in secondary schools will be used as an exemplar. Several phases of work are planned, early phases focus on using data science methods to refine clinical phenotypes, and using characterise health trajectories in relation to environmental factors (describing the “exposome”). Later phases will explore the potential for linked data to be used as a platform for enhancing the efficiency of randomised controlled trials of individually or cluster-randomised (at school or hospital-level) interventions.

Dr Watjana (Waty) Lilaonitkul. Under HDR-UK’s priority area of ‘Actionable Analytics – Unlocking the longitudinal clinical phenome’, Waty’s work will focus on applying artificial intelligence (AI) approaches for medical treatment and diagnosis using digital data stored in anonymised electronic health records (EHR). An AI approach has the potential to support medical decisions by refining its ability to cross-reference complex data, find correlations and consider seemingly unrelated external factors that physicians and researchers may not immediate deem relevant. The model’s performance will be evaluated against physician and existing state-of-the art model performances by considering the timeliness of the prediction and other metrics that quantify accuracy and discrimination.

Dr Tom Lumbers. Tom’s research seeks to reclassify heart failure patients according to underlying causal mechanisms by integrating diverse patient data from genotypes to disease outcomes. He will analyse EHR enriched-genomic data from HDR-UK London, UK Biobank, and the HERMES Consortium, using emerging genome bioinformatic tools and unsupervised learning approaches. The overall aim is to identify and validate molecular pathways as candidate targets for therapeutic intervention and to inform the design of clinical trials.