The University of Bath is inviting applications for a fully funded PhD studentship based in the Department of Computer Science and under the supervision of Dr Mohammad Golbabaee (lead supervisor) https://mgolbabaee.wordpress.com/
and Prof Peter Hall (co-supervisor) https://researchportal.bath.ac.uk/en/persons/peter-hall
The position is an opportunity to conduct cutting-edge research at the intersection of computational medical imaging and machine learning. This is a truly interdisciplinary studentship which runs in close collaborations with outstanding scholars, clinicians and bioengineers in industry (GE Healthcare) and academia (Technical University of Munich and the University of Edinburgh). You will be strongly supported by an expert supervisory team to generate impact, with the potential to influence practitioners in clinical medicine and neuroscience through introducing the next generation imaging tools that could be applied for early detection of cancer and/or neurodegenerative disorders such as dementia or MS.
With recent advances in machine learning, we are witnessing a new wave of AI driven approaches emerging to uncover the physical models behind many complex systems in modern applications. In this project, we particularly aim at developing machine learning solutions for Quantitative Magnetic Resonance Imaging (Q-MRI).
Rather than computing a single MRI image with only contrast information (i.e. qualitative MRI), in Q-MRI physicists are interested in computing the fundamental properties of tissues such as relaxation times, proton densities, diffusion and perfusion parameters, from a time-series of magnetization images. Despite proven a more powerful diagnostic tool than qualitative MRI, Q-MRI never became clinically applicable because of its inconveniently long acquisition time.
Leveraging on Compressed Sensing [1,2] and the novel MR Fingerprinting (MRF) [3-5] paradigms, this studentship will address this challenge and make the Q-MRI runtime feasible in clinical setups. This will involve building new computational models based on Deep Learning that are capable of accurate and real-time MRI quantification given severely under-sampled images (resulted by ultrafast acquisitions). Further, the training datasets for medical imaging are scarce . Therefore, this research will require designing efficient neural network architectures and advanced training strategies to overcome this fundamental challenge for learning the complex physical dynamics behind Q-MRI.
You will be an enthusiastic and self-motivated person who meets the academic requirements for enrolment for the PhD degree at The University of Bath. You will have a 1st class or 2:1 degree (or equivalent outside of the UK) in Computing, Mathematics, Engineering or a related subject. You must have a strong interest in computational medical imaging. Strong mathematical background and experience in signal/image processing, as well as strong programming experience in Python and MATLAB are required. Previous experience in deep learning and computer vision is highly desirable. Good team-working and communication skills are essential.
Informal enquiries should be directed to Dr Mohammad Golbabaee, [email protected]
Formal applications should be made via the University of Bath’s online application form: https://samis.bath.ac.uk/urd/sits.urd/run/siw_ipp_lgn.login?process=siw_ipp_app&code1=RDUCM-FP01&code2=0014
Please ensure that you quote the supervisor’s name and project title in the ‘Your research interests’ section.
More information about applying for a PhD at Bath may be found here: http://www.bath.ac.uk/guides/how-to-apply-for-doctoral-study/
Anticipated start date: 28 September 2020.
UK and EU candidates applying for this project will be considered for a University Research Studentship which will cover UK/EU tuition fees, a training support grant of £1,000 per annum and a tax-free maintenance allowance at the UKRI Doctoral Stipend rate (£15,009 in 2019-20) for a period of up to 3.5 years.
We welcome all-year-round applications from self-funding candidates and candidates who can source their own external sponsorship.
 Baraniuk, R. G. "Compressive sensing." IEEE signal processing magazine 24.4 (2007).
 Lustig, M., et al. "Compressed sensing MRI." IEEE signal processing magazine 25.2 (2008): 72.
 Ma, D., et al. "Magnetic resonance fingerprinting." Nature 495.7440 (2013): 187.
 Golbabaee, M., et al. "Coverblip: accelerated and scalable iterative matched-filtering for magnetic resonance fingerprint reconstruction." Inverse Problems (2019).
 Golbabaee, M., et al. "Geometry of deep learning for magnetic resonance fingerprinting." IEEE International Conference on Acoustics, Speech and Signal Processing (2019).
 Mardani, M., et al. "Neural proximal gradient descent for compressive imaging." Advances in Neural Information Processing Systems. 2018.