PhD in Deep learning for clinical decision support

University of Edinburgh, UK
Location: 
Edinburgh, UK
Job Type: 
Full Time
Closing Date: 
Saturday, December 30, 2017

Short description:

In collaboration with Toshiba Medical Visualization Systems, this project will develop methods, algorithms and theory that realize decision support systems via deep representation learning mechanisms based on both imaging and non-imaging clinical data.   This requires clinically meaningful statistical distances to be learned, taking advantage of both annotated and weakly (or non) annotated data – the latter being plentiful and easier to source.  Such semi-supervised approaches to learning are increasingly sought-after in many other fields (e.g. in multimedia where both video, audio and text are available).  Of particular interest is how one can balance the contributions of different information sources (imaging vs. non-imaging sources) and how to deal with missing or incomplete observations.  The project will draw inspiration from generative methods to build robustness and unearth the importance of each source.  For the system to be credible it should be able to present its reasoning by identifying the most significant information in each case, and exposing its degree of confidence.  Consequently, this project will also consider how generative methods can provide confidence estimates.

To achieve this, we are looking for an enthusiastic and strongly motivated student to join our team. He/she will have the opportunity to collaborate with Toshiba researchers in Edinburgh and other collaborators throughout the world.

Further Information: 

For additional information on the research we do and the supervisor please see http://tsaftaris.com ; informal inquiries can be addressed to Dr Tsaftaris, S.Tsaftaris@ed.ac.uk, with a CV attached.

Funding: 

This position can be fully funded for 42 months (3.5 years) and is open to UK nationals and EU nationals (if they have resided for more than 3 years in the UK).

 

Candidate profile: 

Candidates should have a Master’s level education (or in exceptional cases an excellent (upper second class and higher) Bachelor’s degree) in electronic/electrical engineering, computer science (informatics), physics or closely related subjects. Any prior experience in medical image analysis is desirable but not necessary.  Prior exposure to machine learning is desirable but not essential. The candidate will be expected to have a high level of analytical and investigative skills. The candidate should have good programming skills (e.g., Python, Matlab) and a solid mathematical background.

Deadline

Thursday, December 31, 2017 but earlier application is highly encouraged.

To apply:

https://www.eng.ed.ac.uk/postgraduate/research/projects/deep-learning-clinical-decision-support-0

About the University and Edinburgh

The University of Edinburgh is considered one of the top universities in the world according to recent rankings. We have excellent faculty working on theory and applications of machine learning and signal processing in healthcare, and a vibrant culture of students and postdocs.  In addition, we are in close proximity to excellent imaging facilities which include preclinical and clinical scanners as well as state-of-the-art dual modality scanners (e.g., PET/MR) and collaborate with several clinicians.

Edinburgh, the capital of Scotland, offers a vibrant professional life, excellent career opportunities and a high quality of life.  It has a beautiful old part, has an abundance of café’s, restaurants and bars, and yearly hosts the Festival and the Fringe, which is the largest arts gathering in the world.