Research Associate in Generative Deep Learning for Synthetising Virtual Patient Populations for In-Silico Trials

We are seeking an ambitious and proactive Research Associate to be part of a multidisciplinary team, focusing on image-based multiphysics modelling of cardiovascular fluid dynamics and device-tissue interactions. The successful candidate will utilise clinical and experimental data to pioneer novel generative AI and geometric deep learning approaches to create synthetic virtual patient cohorts from multimodal data. This role involves developing advanced algorithms and high-throughput workflows for crafting virtual populations and simulation-ready computational anatomy models, integrating tissue microstructure properties where relevant. The role requires applying innovative techniques to large, real-world multimodal datasets, including clinical trials and population imaging studies.

What you’ll need

Applicants should have a PhD (or nearing completion) or equivalent in computational imaging and deep learning, and an understanding of applied mathematics, focusing on algorithm design and analysis. Proficiency in modern ML techniques, including geometric deep learning, diffusion models, and neural networks for multimodal image analysis will be essential, as well as expertise in Python and C/C++ for scientific computing, and in ML/DL frameworks like TensorFlow, PyTorch, Keras, and Scikit-learn. A developing publication profile will be advantageous.

As this role involves research at a postgraduate level, applicants who are not an EEA national or a national of an exempt country and who will require sponsorship under the Skilled Worker route of the UK Visas and Immigration’s (UKVI) Points Based System in order to take up the role, will be required to apply for an Academic Technology Approval Scheme (ATAS) Certificate and will need to obtain this prior to making any official visa application UKVI.