Precision medicine approaches are needed to revolutionise management an improve outcomes for patients with inflammatory bowel disease (IBD), a chronic immune mediated inflammatory disease of the gastrointestinal tract without a medical cure. Although there are multiple different classes of treatments of IBD, only few patients achieve sustained remission. The development of robust biomarkers to predict response to these therapies would allow individual IBD patients to be fast-tracked to the right biological agent, such that improved outcomes can be achieved at the earliest time point possible.
In this PhD project the student will test the hypothesis that features collected as part of routine clinical care may be used to segregate patients to groups with different prognosis allowing to identify early those patients who may have a less than favourable clinical trajectory. The student will also investigate the genetic susceptibility for IBD. For this, the PhD candidate will develop semi-supervised and unsupervised clustering algorithms based on complex machine learning approaches to segregate patients with IBD in groups of different prognostic trajectory. The project will use data from three cohorts (GSTT IBD, KCH IBD and IBD Bioresource, adding a total of ~200,000 patients with IBD). Dr Iniesta will support the student’s training and career development, will provide them with expert guidance on prediction modelling, machine learning, general statistical procedures, scientific writing and scientific dissemination approaches. Dr Pavlidis, an expert clinician and scientist on chronic inflammatory gastrointestinal conditions, will support the student’s learning on the clinical and methodological angles of the PhD project.