Machine Learning for AML Registry
PhD project: Machine Learning & Real-World Data Modelling in the UK Acute Myeloid Leukaemia (AML) National Registry
Overview
Warwick Medical SchoolLink opens in a new window invites applications for a fully funded 4-year PhD applying advanced data science and artificial intelligence methods to one of the UK’s most comprehensive national datasets for acute myeloid leukaemia (AML).
Applicants must be British nationals or Residents (Home students).
Supervisors:
- Professor Charles Craddock (Warwick University)
- Doctor Priyanka Mehta (Bristol)
- Professor Christopher Smith (Warwick University)
How to Apply
Apply hereLink opens in a new window
UK (British nationals/Residents) Applicants only
Provide your CV, transcripts, and Personal Statement.
- Deadline : 13 April 2026
- Provisional interview date: w/c 27 April 2026
- Start Date: October 2026
Project Summary
Acute myeloid leukaemia (AML) is an aggressive blood cancer affecting approximately 3,000 people annually in the UK, with survival remaining poor, particularly in older patients. The MyeCare Registry is the first UK leukaemia registry to integrate detailed clinical outcomes with whole genome sequencing data from patients across the NHS, providing an unprecedented national real-world dataset.
This PhD will develop and rigorously evaluate machine learning methods to model clinical and genomic registry data, with the aim of improving risk stratification, understanding treatment response and survival, and generating robust real-world evidence to support future clinical decision-making and trial optimisation.
In addition, there are opportunities to apply transferable methods to the Out of Hospital Cardiac Arrest Outcomes (OHCAO) registry, a national dataset of over 380,000 cases, to explore intervention modelling and population-level risk prediction.
The work will involve:
-
Data preparation and integration of clinical and genomic registry data
-
Predictive modelling using machine learning and survival analysis methods
-
Model evaluation and interpretation, including comparison with traditional statistical approaches
-
Identification of clinically relevant patterns related to disease progression and treatment outcomes
The aim is to develop robust and interpretable models that can generate new insights into AML outcomes using large-scale registry data.
This studentship is ideal for candidates with strong quantitative skills with background from:
- Artificial Intelligence & Machine Learning
- Bioinformatics
- Mathematics
- Data Science
- Computer Science
You will have experience and strong interest in data analysis and wish to apply AI to real-world and high-impact healthcare challenges.
As a PhD candidate, you will design transferable machine learning and AI models, work with largescale biomedical and registry data sets, collaborate with clinicians, bioinformaticians, data scientists and trialists with a goal to contribute to sustainable and ethical AI in healthcare.