University of Leeds Featured PhD Programmes
University of Kent Featured PhD Programmes
Newcastle University Featured PhD Programmes
Catalysis Hub Featured PhD Programmes
University College London Featured PhD Programmes

MRC Precision Medicine DTP: Machine learning for automated assessment of CT angiography imaging in stroke

Project Description

Stroke is a major cause of death and disability and with >100,000 people affected annually in the UK, is estimated to cost our economy £25 billion per year. Most strokes (~80%) occur following a sudden blockage of an artery supplying the brain. The most effective treatment for many of these patients is emergency thrombectomy where the arterial blockage (usually blood clot) is pulled from the affected vessel using a minimally invasive technique.1

Rapidly identifying the correct patients for thrombectomy relies on CT angiography (CTA) to demonstrate the location of arterial blockage. CTA provides excellent anatomical representation of the major arteries supplying the brain. In addition, CT can provide prognostic information by assessing various imaging biomarkers including collateral blood supply to the brain and indicators of brain frailty. In the context of precision medicine, this information can be combined with other clinical metadata to more accurately predict individual patient outcome after stroke and thereby assist clinical decision making.2 Interpretation of CTA generally requires expert radiological knowledge but this expertise is not always available, especially outside of normal working hours, or remote from major treatment centres. As such, there can be significant delays in the interpretation of CTA. Unfortunately, delays in stroke care make poor outcomes more likely.3

Machine learning offers a method of developing automated systems using artificial intelligence for the rapid, precise and repeatable assessment of medical imaging and for combining imaging and clinical metadata in predictive models.4 But there are hardly any commercially available computer vision solutions for assessing imaging in stroke and none can interpret CTA for the purposes of thrombectomy decision-making.

Development of state-of-the-art computer vision systems requires labelled imaging datasets that are representative for the imaging features of interest. We have immediate access to large CTA datasets acquired within clinical stroke trials. These datasets include expert neuroradiologist interpretation and a wealth of clinical meta-data. Therefore, we have validated and labelled these datasets for several clinically relevant imaging and non-imaging characteristics that will act as ‘ground truth’, and with reference to current evidence will provide the basis for predictive modelling. This presents an exciting opportunity to develop a machine learning system to output clinically meaningful information. Ultimately, this tool will provide valuable treatment decision support for front-of-house clinicians; chiefly to determine individual patient suitability for thrombectomy and to estimate the likelihood of good versus poor outcome should thrombectomy be successful.

-Develop an innovative system using machine learning for the automated and real-time assessment of CT angiography in stroke.
-Work within a vibrant interdisciplinary clinical-academic-industrial research collaboration benefitting from what individual environments have to offer.
-Help deliver the next generation of commercially viable expert medical image processing software for rapid and precise diagnostic support

Training outcomes
-Develop proficient skills in multidisciplinary team working, software engineering, image processing, data analysis and statistics, clinical translation, interpretation and dissemination of results through various media as expected at PhD level.
-Employ state-of-the-art machine learning (such as convolutional neural networks) in the context of computer vision to develop an automated image analysis and treatment decision support system for CTA.
-Interpret various imaging biomarkers and apply this knowledge for diagnosis and outcome prediction in stroke.
-Use probability theory and predictive modelling to develop both diagnostic and prognostic capabilities in the system and thereby offer treatment decision support to the end user.
-Present work at major national and international meetings and publish results in high impact peer-reviewed journals.
-Understand the process for developing, testing and delivering commercially viable image processing software.
-Work effectively within an industrial-academic-clinical collaboration and gain insight into future career options across disciplines.
This MRC programme is joint between the Universities of Edinburgh and Glasgow. You will be registered at the host institution of the primary supervisor detailed in your project selection.

All applications should be made via the University of Edinburgh, irrespective of project location:

Please note, you must apply to one of the projects and you should contact the primary supervisor prior to making your application. Additional information on the application process if available from the link above.

For more information about Precision Medicine visit:

Funding Notes

Start: September 2019

Qualifications criteria: Applicants applying for a MRC DTP in Precision Medicine studentship must have obtained, or will soon obtain, a first or upper-second class UK honours degree or equivalent non-UK qualifications, in an appropriate science/technology area.

Residence criteria: The MRC DTP in Precision Medicine grant provides tuition fees and stipend of at least £14,777 (RCUK rate 2018/19) for UK and EU nationals that meet all required eligibility criteria.

Full eligibility details are available: View Website

Enquiries regarding programme:


1. Rodrigues FB, Neves JB, Caldeira D, Ferro JM, Ferreira JJ, Costa J. Endovascular treatment versus medical care alone for ischaemic stroke: systematic review and meta- analysis. BMJ. 2016;353:i1754
2. Liebeskind DS, Malhotra K, Hinman JD. Imaging as the Nidus of Precision Cerebrovascular Health: A Million Brains Initiative. JAMA neurology. 2017;74:257-258
3. Saver JL. Time is brain - quantified. Stroke. 2006;37:263-266
4. Chartrand G, Cheng PM, Vorontsov E, Drozdzal M, Turcotte S, Pal CJ, et al. Deep Learning: A Primer for Radiologists. Radiographics. 2017;37:2113-2131

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully

FindAPhD. Copyright 2005-2019
All rights reserved.