We seek a Ph.D. student to develop next-generation, low-cost, ultra-thin endoscopes that can image diseases including cancer in hard-to-reach areas of the body such as the pancreas and ovaries.
Background: Cancers that occur deep within the body are difficult to detect and treat due to physical inaccessibility: ovarian cancer has a 50% 5-year survival rate while for pancreatic cancer this is just 1%. To address this, we are developing ultra-thin optical fibre endoscopes that will use nano-structured meta-surfaces to implement advanced imaging modalities that can fit inside a needle, go deep inside the body and ‘see’ disease much better than conventional cameras. However, the images produced are often distorted as the light travels through the fibre so the full potential of these devices may only be realised when combined with advanced computational techniques for image reconstruction. We are therefore assembling an interdisciplinary team, led by Dr. George Gordon as part of a UKRI Future Leaders Fellowship, and are looking for a highly motivated and creative Ph.D. student to work on the AI and computational imaging theme of the project.
Aim: The aim of this project is to explore the application of cutting-edge AI techniques, such as image transformers and graph neural networks, to co-design optical metasurfaces and image reconstruction algorithms for optimal performance. It may also involve using reinforcement learning for real-time reconstruction and physics-informed neural networks to derive explainable models. You will work closely with other members of the lab who are designing and building nano-structured metasurfaces on optical fibres and also with clinical collaborators, with the chance to apply your techniques to real-world clinical data we will soon collect. The role would be suitable for someone passionate about computing, mathematics with a desire to develop skills in AI.
What we offer:
- A chance to learn new skills: artificial intelligence (e.g. transformers, graph neural networks, reinforcement learning), optics, nanotechnology, clinical applications
- A world-class multidisciplinary research environment, spanning nano-technology to AI to clinical medicine, working closely with other members of the team
- A supportive environment for researchers as signatories of the Researcher Development Concordat (www.vitae.ac.uk/policy/concordat). See also our lab charter (http://www.georgesdgordon.com/?page_id=584)
- The opportunity to produce high-quality publications
- Travel to visit international partners and attend conferences
- 3.5-years funding including fees (UK/home-fee status applicants preferred but international applicants can apply for scholarship to top-up – please contact to discuss)
What you should have:
- 1st or 2.1 honours undergraduate or master's degree in Engineering, Physics, Computer Science, Mathematics or similar
- Some coding abilities in any language (e.g. Python, MATLAB, C/C++)
- Excellent communication skills
- The ability to work collaboratively in a team across disciplines
The project will be based in the lab of Dr. Gordon (OPTIMlab) in Engineering, which has 3 main research themes: novel optical materials, new hardware and algorithms for optical imaging, and clinical translation (more at www.georgesdgordon.com). You will be supported by colleagues with strong skills in building optical systems, coding and data analytics/processing.Informal enquiries may be sent to [Email Address Removed]. Please note that formal applications sent directly to this email address will not be accepted. Applications should include a CV and covering letter detailing relevant experience and why they are suitable for the role.
Our University is a supportive, inclusive and caring community and we encourage applications from a diverse range of backgrounds. The Faculty of Engineering was the first in the UK to be awarded an Athena SWAN Gold Award, in recognition of our commitment to supporting and advancing women’s careers in Engineering.
Start date: October 2024.
Closing date: 31 January 2024
Continue with Facebook
