Lei Clifton
BSc (Hons), MSc, PhD
Associate Professor
- Official Fellow in AI & ML, Reuben College
- Programme Director of the MSc in Applied Digital Health
AI and statistics for real-world healthcare applications
I work at the intersection of AI and medical statistics, bringing over 20 years of research experience. My research spans machine learning, foundation models (e.g. large language models), and classical statistics, with a focus on developing tools for real-world clinical settings. I collaborate widely, including with the Computational Health Informatics group in the Department of Engineering Science, where I also hold an affiliation.
I am the Programme Director of the MSc in Applied Digital Health, where I am also module leader and lecturer, working alongside the Academic Directors, Profs. John Powell and Catherine Pope.
After studying engineering and machine learning, I did my postdoctoral training at the Department of Engineering Science (2009-14), before training as a medical statistician under Prof. Doug Altman at the Centre for Statistics in Medicine (2014-18). Subsequently, I was team leader in the Nuffield Department of Population Health (2019-24), where I led a programme of research in translational epidemiology.
When at home, I can be found painting watercolours, practising yoga, making noise on the violin with friends, or playing board games with my husband and two boys.
Recent publications
Assessing the Importance of Variation in Diagnostic Coding Among the Three Countries in the UK Biobank.
Journal article
Clifton L. et al, (2026), Learn Health Syst, 10
Democratising Clinical AI through Dataset Condensation for Classical Clinical Models
Preprint
Thakur A. et al, (2026)
Digital morphine: why AI scribes are symptomatic relief for a broken system
Journal article
Segal B. et al, (2026), BMJ Digital Health & AI, 2, e000030 - e000030
Cardiac health assessment across scenarios and devices using a multimodal foundation model pretrained on data from 1.7 million individuals
Journal article
Gu X. et al, (2026), Nature Machine Intelligence, 8, 220 - 233