Investigating the use of Bayesian prediction methods in precision medicine approaches for Type 2 diabetes - PhD in Medical Studies (Research England DTP)

Type 2 diabetes is a major public health problem that is responsible for around 90% of all diabetes cases. There are many factors that underlie not only an individual’s risk of developing diabetes, but also their subsequent progression and the efficacy of different treatment options. Precision medicine approaches aim to improve patient care by allowing doctors to use personal information to better understand disease risk in individual patients, and also to inform optimal treatment choices.

This PhD will develop novel Bayesian statistical approaches for tackling various key aspects of effective precision medicine, particularly focused on prediction and treatment selection models (e.g. Jones et al. 2016, Dennis et al. 2018, Dennis et al. 2018). The student will develop new methods to overcome the challenges of modelling complex human data, including key statistical issues such as overfitting. The student will also explore Bayesian adaptive learning techniques to improve predictions for assessing the efficacy of treatments and progression of individual patients that can be updated over time as new data becomes available.

The student will have the opportunity to learn and develop cutting-edge statistical and machine learning methodologies, and work with a world-renowned diabetes research team. They will use existing clinical research data on over 400,000 real world UK patients with Type 2 diabetes as well as data from 1000s of individuals who have participated in randomised drug trials.

The project would suit someone with a background in statistics, machine learning or mathematics. Experience in handling complex real-world data sets is desirable. The student will join a vibrant, diverse and world class interdisciplinary research team, including geneticists, biological scientists, mathematicians, computer scientists and clinicians, all studying aspects of diabetes and related conditions.