Imaging science is an active branch of applied mathematics. This project focuses on segmentation which is one of the challenging problems in imaging. Although there is a vast literature on variational models and their modern counterpart of deep learning algorithms. There remain many challenging areas in which current techniques are not yet sufficient. This project is concerned with one such area: segmentation of regions of interests in microscopic images. The field is growing rapidly as new technologies and modalities are constantly driving the development.

This imaging project is an opportunity to undertake one of our new and exciting cross-disciplinary projects lying at the interface of mathematics and cancer medicine. A core technology in the latter is based on intelligence use of microscopic images in different modalities. The initial stage of the project will be on familiar with recent works (in-house methods and codes) from the group and the next stages are to explore deep learning methods that merge natural language input and mathematical algorithms to improve the state of the are algorithms for various segmentation tasks from microscopic images. Ideas from the popular adaptive transformers and generative models as well as links of language models to distance functions will be considered and explored.

The candidate does not need to have any knowledge in cancer diseases or medicine but will need strong mathematical knowledge through a degree in mathematics, or in computer science / physics / engineering with essential mathematics components, as well as reasonable programming skills. Training in these topics will be given.

The successful candidate will have the opportunity to work in an active research group, with research scientists, PhD students, Postdocs, industrial and NHS collaborators.

The successful student will get a thorough training in Mathematical Imaging (e.g variational models and diffeomorphic maps), Artificial Intelligence (neural networks and transfer learning), Biomedical Research (Multimodal imaging, transcriptomics and H&E) and gain practical problems solving skills that are highly Valued in both Academia and Industries.
Overall, this fully funded studentship is attractive because the topic under study is interesting and modern, and a candidate has the flexibility to focus on all or just one of the key components: mathematics, AI and biomedical analysis in terms of leading research.

Further information

Please visit Professor Ke Chen's website