Primary Supervisor – Dr YingLiang Ma Background Cardiovascular disease is the leading cause of death in the UK. As of 2023, over 1.6 million people in the UK have been diagnosed with heart arrhythmias, and 900,000 with heart failure. Medical imaging techniques, such as CT, MRI, and X-ray, are vital for diagnosing and guiding the treatment of cardiovascular diseases. The goal of this project is to develop deep learning-based models that can accurately segment blood vessels and heart chambers from X-ray fluoroscopic videos, as well as from clinical CT and MRI scans. Additionally, we aim to reconstruct 3D models of blood vessels and heart structures using this data. You will collaborate with a multi-disciplinary team, which includes cardiologist consultants from St. Thomas Hospital London and University Hospital Coventry and Warwickshire. The project is partnered with King’s College London and Fudan University, China. This project is linked to an EPSRC-funded project titled "3D Hybrid Guidance System for Cardiac Interventional Procedures. Research methodology You will use a large medical image database to train deep learning algorithms and are expected to contribute to the development of data labeling software. Additionally, the algorithms must be robust to noisy medical images and incomplete data. To reduce the burden of data labeling, particularly for X-ray fluoroscopic image sequences, semi-supervised or weakly supervised learning methods will be employed. For 3D reconstruction, Neural Radiance Field will be explored to reconstruct 3D anatomical models from limited X-ray views. Training You will be based at the Vision & Graphics Lab within the School of Computing Sciences, which specializes in deep learning for medical image analysis and neural fields for 3D reconstruction. This position provides an opportunity to collaborate with scientists from partner institutes around the world. You will also receive specialized training in high-performance computing and the use of GPU clusters. Funding Details Additional Funding Information This PhD project is in a competition for a Faculty of Science funded studentship. Funding is available to UK applicants and comprises ‘home’ tuition fees and an annual stipend of £19,237 (for a maximum 3 years) Closing Date: 27 November 2024 (at 11.59 pm)