PhD in Mechanistic and Interpretable AI for Personalized Eczema Severity Forecasting at Imperial College London Imperial College London in United Kingdom

Imperial College London is advertising a PhD studentship in mechanistic and interpretable AI for personalized eczema severity forecasting through the MultiSci MRC DTP/iCASE programme.

The project is led by Professor Reiko Tanaka with Professor Adnan Custovic as co-supervisor, and includes collaboration with Pierre Fabre Laboratories. The research sits at the intersection of computer science, medical science, biology, and statistics, with a strong emphasis on machine learning, Bayesian modelling, time-series forecasting, computer vision, and analysis of biomedical data.

The project aims to build AI tools that predict eczema flare-ups 1–3 days ahead using smartphone photos, skin barrier measurements, and biological markers such as bacteria and lipids. A key feature is mechanistic interpretability, meaning the model should explain why a flare risk is rising rather than acting as a black box. The work is interdisciplinary and clinically relevant, bridging dermatology, systems biology, and AI.

Eligibility is restricted to UK home fee status applicants only. The post specifically says applicants should have a strong maths background. The Imperial project page notes that applications are reviewed on a rolling basis and the position will be filled once a suitable candidate is identified.

Funding is through an iCASE PhD studentship with an industry partner and includes time spent working with the company. The project page also indicates tuition fees are covered at the UKRI rate for eligible students.

To apply, review the project details, choose this project in the studentship application form, and submit your application through Imperial College London as soon as possible.