Expires on: 06/30/2024
Modern deep learning techniques achieve human-like performance in many medical image analysis tasks, including the identification of anomalous tissue/pathology from medical scans. To be trained, these techniques typically require large image datasets with pixel-level annotations provided by medical experts. However, obtaining reliable annotations is very difficult (due to the intrinsic nature of the task, especially for rare/complex pathologies) and highly time-consuming. This severely hinders the development and deployment of AI into clinical practice, despite its huge potential.
This PhD Scholarship will focus on designing novel approaches that require less detailed/reliable annotations but are still capable of producing highly accurate results. These approaches will include training models through weak supervision (i.e. leveraging only coarse annotations provided by the experts) and incorporating noise-robust learning strategies (i.e. accounting for the presence of unreliable annotations). We expect that many high-impact publications will be generated during the project, to be presented both in computer science-related venues (e.g. CVPR, NeurIPS, MICCAI) as well as at medical conferences (e.g. ISMRM, ESMRMB).
The PhD candidate will work in an exciting international environment in the heart of the City of London. They will join the School of Science and Technology at City, University of London (member of the Alan Turing University Network) and the CitAI Research Centre (which features academic staff with extensive expertise in machine learning for healthcare). They will also be able to exploit the power of Hyperion, City’s High-Performance Computer.
This Scholarship will be carried out in collaboration with St George’s, University of London (which is merging with City University). The candidate will have access to St George’s highly valuable clinical datasets (e.g. MRI of patients with brain tumours, brain injury, diseases of aging) as well as supervision from leading biomedical researchers with strong links to radiology. Consequently, the research outputs of this Scholarship will have potential for impact in clinical practice.
Benefits
The Scholarship includes:
- A competitive annual bursary for 3 years (£21,000/year)
- Full tuition fees for UK/Home Students. Partial fee coverage for European/Overseas Students
- The opportunity to earn up to £4,300/year through a non-compulsory teaching assistantship
- Over £4000 to participate to conferences and training
Eligibility:
The studentships will be awarded based on outstanding academic achievement and the potential to produce cutting-edge research. Prospective applicants must:
- Hold a good honours degree (no less than a second-class honours degree or an equivalent qualification) in an appropriate subject
- Knowledge of modern machine learning techniques for computer vision and experience with coding in Python is beneficial (but not a strong requirement)
- Applicants whose mother tongue is not English must meet any one or a combination of the following:
- A minimum IELTS average score of 6.5; with a minimum of 6.0 in each of the four components
- The award of a Masters’ degree, the teaching of which was in English from an English-Speaking Country