Please find the details of the available projects for AI and Data Science studentships outlined below. A full list of our available projects can be downloaded here. Contact details for the lead supervisor can be found by clicking on their name.
For details on how to apply, please head back to our How to Apply page. Our application form can be found here.
Developing and using intelligent machine learning tools to re-classify kidney disease based on novel molecular fingerprints
We will use advanced methods from machine learning in application to genomic medicine, see e.g. https://seis.bristol.ac.uk/~enicgc/software.htm. Though the applications areas will be broad-based, a particular focus will be chronic kidney disease and nephrotic syndrome, in collaboration with Prof. Moin Saleem (University of Bristol).
Supervisor: Dr Colin Campbell.
Mental health data science in rich longitudinal population cohorts
Online social networks provide a wealth of information about our mental states, but the true value can only be realised when we link online measurements of mental health to gold-standard measures collected offline. This project will use social media data collected in large population cohort studies to validate digital signatures of mental health.
Supervisor: Dr Claire Haworth.
Next-generation clinical bacterial diagnostics with MALDI mass spectrometry learnt from vast historical data
In infections such as sepsis, each hour without working antibiotics leads to a 6% increase in mortality. Analysing bacterial cultures with MALDI-ToF mass spectrometry leads to robust species identification, but there is an urgent need to address insensitivities in the informatics pipeline to speed up results and determine antibiotic resistance.
Supervisor: Professor Andrew Dowsey.
Novel data-driven pathways to impact from the causal network of human health
We can infer how a medical intervention will ripple through a causal network of human phenotypes, potentially influencing many health outcomes in complex and unexpected ways. This project will build methods and software that enables the public and policy makers to interact with the causal network, helping us learn to make better interventions.
Supervisor: Dr Gibran Hemani.
Machine Learning for Prediction in Alzheimer’s Disease: Identifying Novel Biologically Valid Diagnostic Categories to Inform Precision Medicine
The project will involve investigating the ability of machine learning algorithms for improving biologically-based classification of cases/controls in Alzheimer’s Disease and will make use the rich phenotype information available in UKBiobank to improve predictions of AD-associated outcomes and other dementia related phenotypes.
Supervisor: Professor Valentina Escott-Price.
Machine-learning guided immune fingerprinting for rapid detection of life-threatening infection in liver cirrhosis
The student will create a computational framework for the analysis of complex biomedical datasets, and develop and apply bespoke machine learning-based approaches that identify biomarker signatures in patient samples to predict the presence of life-threatening infections in individuals with cirrhosis of the liver.
Supervisor: Dr Matthias Eberl.
Data science and predictive analytics to improve treatment for patients with Type 2 diabetes
The availability of massive, individual level, longitudinal data from trials and routine practice offers new opportunities for diabetes care. Using data science and machine learning we will develop, validate, and bring into clinical care a decision aid which balances for an individual likely efficacy and side effects of 8 possible treatments.
Supervisor: Professor Andrew Hattersley.
Determining epilepsy subtypes: A Data Science Approach
Epilepsy is a serious condition affecting 500,000 people in the UK alone. Diagnosis and prognosis of epilepsy is challenging, with easily identifiable features from clinical imaging the current state of the art. Here we will develop data science and mathematical tools to reveal new markers of epilepsy from these routinely acquired clinical data.
Supervisor: Professor J.R. Terry.
Using birth cohorts to understand the impact of urban green space on child health and wellbeing
This interdisciplinary project will investigate the impact of the environments in which children grow up on their health and wellbeing. It will use systematic review, geographical and epidemiological methods, and two UK birth cohorts to investigate the role of urban green space in shaping child and adolescent physical and mental health.
Supervisor: Dr Benedict Wheeler.