Multi-state survival models allow rich insights into complex disease pathways, where a patient may experience many non-fatal/intermediate events, and we wish to the investigate covariate effects for each specific transition between two states, not just for example, on the first event, or a terminal event . With the growth in availability of electronic health records (EHRs), there is substantial opportunities to build biologically complex, yet plausible, disease risk models, which take into account an entire disease trajectory, rather than just a single event of interest (such as a myocardial infarction or stroke). The size of such datasets also opens up possibilities to utilise data-hungry machine learning methods in new fields of application.
The aim of this PhD is to bring together established biostatistical approaches, with emerging machine learning techniques to answer important clinical questions in the field of cardiovascular disease (CVD). The PhD has the following potential goals:
• To conduct a review of survival analysis methods applied to CVD risk prediction
• To compare statistical and machine learning (ML) approaches to survival analysis/risk prediction through extensive simulation studies, including for example, flexible parametric survival models , penalisation approaches, neural networks , random forests
• To develop and extend existing survival analysis methods to incorporate machine learning techniques, for example combining flexible parametric survival models with an artificial neural network. There will be the opportunity to develop and release open-source user-friendly software packages associated with the new methods.
• To extend the developed approach to a general multi-state setting, allowing any number of states and transitions 
• To develop a prediction model using big data/EHRs in cardiovascular disease and cancer, for example using data from the first whole country cardio-oncology research platform (VICORI) to predict multimorbid events in a stable coronary artery disease population
The PhD student will join the Biostatistics Research Group, which is an internationally regarded hub of biostatistical methods development. The group is home to leading researchers in the field of survival analysis (Michael Crowther, Michael Sweeting, Mark Rutherford and Paul Lambert), whose research focus on development of methods for the analysis of complex survival data, motivated by applications to electronic health records. There will be the opportunity to attend specialist training courses such as ‘Parametric competing risks and multi-state survival models’ (taught by Michael Crowther and Paul Lambert at the Swiss Winter Epidemiology School), and attend and present at international conferences, such as the International Society for Clinical Biostatistics’ annual conference.
Applicants are required to hold/or expect to obtain a data science related UK Bachelor Degree 2:1 or better (e.g. Computer Science, Bioinformatics, Biostatistics), and preferably also a similar MSc qualification. The University of Leicester English language requirements apply where applicable: https://le.ac.uk/study/research-degrees/entry-reqs/eng-lang-reqs
How to apply:
You should submit your application using our online application system: https://www2.le.ac.uk/research-degrees/phd/applyphd
Apply for a PhD in Health Sciences Research
In the funding section of the application please indicate you wish to be considered for a CLS HDRUK Studentship
In the proposal section please provide the name of the supervisor and project you want to be considered for – please list both your first and second choices.
Application enquiries to [email protected]
 Putter H, Fiocco M, Geskus RB. Tutorial in biostatistics: competing risks and multi-state models. Stat Med 2007;26(11):2389–2430.
 Royston P, Parmar MKB. Flexible Parametric Proportional Hazards and Proportional Odds Models for Censored Survival Data, with Application to Prognostic Modelling and Estimation of Treatment Effects. Stat Med 2002;21(15):2175-2197.
 Ripley RM, Harris AL, Tarassenko L. Non-linear survival analysis using neural networks. Stat Med 2004;23:825-842.
 Crowther MJ, Lambert PC. Parametric multi-state survival models: flexible modelling allowing transition-specific distributions with application to estimating clinically useful measures of effect differences. Stat Med 2017;36(29):4719-4742.