Deep Learning Models for Dissolution Dynamics of Amorphous Formulations in Patient-Specific Gastrointestinal Geometry

About the Project

Project description:

Current tools for predicting oral drug performance, such as the TIM-1 system and computational fluid dynamics (CFD) models, provide valuable insight under controlled in vitro conditions but cannot capture variability in gastric geometry and motility across patients. As a result, predictions often fail to reflect in vivo performance, particularly for complex formulations such as amorphous solid dispersions. This project will develop geometry-conditioned neural operators that learn from CFD simulations of drug dissolution and precipitation in patient-specific gastrointestinal geometries. By combining physics-based simulation with deep learning, the framework will enable accurate, patient-level prediction of oral drug performance.

This PhD studentship is a part of CEDAR. CEDAR is an 8.5-year programme funded by an EPSRC Centre for Doctoral Training (CDT) in cyber-physical systems for medicines manufacturing. Through this project, PhD researchers will have a unique opportunity to collaborate with industry leaders to build a comprehensive toolkit for digital and advanced medicine manufacturing processes.

Research challenges addressed in this project:

This project bridges AI, computational fluid dynamics, and pharmaceutical science to revolutionise how we predict oral drug performance. You will develop geometry-conditioned deep learning models that learn how drug dissolution and precipitation behave inside patient-specific gastrointestinal geometries derived from MRI or CT data. By combining machine learning with physics-based simulations, you’ll create fast models on patient-centric drug dissolution dynamics. This project offers the opportunity to work at the cutting edge of pharma innovation, geometry-aware AI, and personalised medicine.

Funding and eligibility:

The studentships cover home tuition fees, research/training costs, and provide a monthly stipend for 4 years (minimum annual stipend of £20,780, tax-free). Part-time opportunities may be available, please contact the skills@cmac.ac.uk for more information.

Entry requirements:

• A minimum of a 2:1 Honours degree (or international equivalent) in chemical engineering, chemistry, computer science, data science, electrical engineering, materials science, mechanical engineering, pharmaceutical sciences, physics, or a relevant science or engineering discipline.
• An MSc is desirable.
• For international students whose first language is not English, an IELTS score of 6.5 (with no less than 5.5 in any element) is required.

To apply: Apply for a PhD: EPSRC CDT in Cyber-Physical Systems for Medicines Manufacturing 2026