AI-powered system to support neuroradiological readings for cerebrovascular diseases

This project is one of 17 four year PhD Studentships funded by Medical Research Scotland (https://www.medicalresearchscotland.org.uk) to be delivered jointly by the named University and External Partner Organisation (EPO). The Studentship will provide the first-class academic and additional training provided by the EPO needed to equip the successful candidate for a science career in an increasingly competitive market.

"AI-powered neuroradiological support system for cerebrovascular diseases from magnetic resonance imaging" to be delivered by the University of Edinburgh [Supervisors: Dr Maria del C Valdés Hernández (Centre for Clinical Brain Sciences, University of Edinburgh), Dr. Beatrice Alex (Edinburgh Futures Institute, University of Edinburgh), Dr William Whiteley (Centre for Clinical Brain Sciences and Usher Institute of Population Health Sciences and Informatics, University of Edinburgh), Dr Miguel O. Bernabeu (Usher Institute of Population Health Sciences and Informatics, University of Edinburgh) and Professor Joanna Wardlaw (Centre for Clinical Brain Sciences, University of Edinburgh)] and Canon Medical Research Europe Ltd (https://research.eu.medical.canon/) [External Partner Organisation supervisor: Dr Alison O'Neil].

Diagnostic and interventional radiology is fundamental to modern healthcare, from getting a fast, accurate diagnosis to planning surgery to cancer care and trauma management. Life-threatening events like strokes require neuroradiological assessment. The UK-wide shortage of radiologists is largely known, and service leaders have stressed that staff shortages are putting patients at risk, with three-quarters saying they cannot guarantee a safe service, as new consultant numbers and overseas and locum appointments are just not enough to plug staffing gaps, situation worsened by the current COVID-19 pandemic. As a result, our radiologists are exhausted and the NHS bill for outsourcing scan reporting continues to climb into the hundreds of millions. This project attempts to address this need by developing an artificial-intelligence powered system to support neuroradiological readings, built upon results from years of neuroimaging and natural-language-processing research.

The student will develop an autonomous system that will use magnetic resonance images with/without previous neuroradiological reports, to automatically generate/make suggestions/detect possible errors in neuroradiological reports with emphasis in reporting cerebrovascular disease indicators, prevalent within neurological-related COVID-19 manifestations; and extract information from a patient’s past medical history to support later clinical decision-making.

ENQUIRIES:

Enquiries should be sent by email to Dr Maria del C Valdés Hernández:

[Email Address Removed]

APPLICATIONS:

Applicants must have obtained, or expect to obtain, a first or 2.1 UK honours degree, or equivalent for degrees obtained outside the UK, in enginnering, informatics or computer science.

Applications are welcome from UK nationals, EU nationals that hold UK settled or pre-settled status and international students.

Applicants are required to provide the contact details of 2 referees (including email addresses) and a supporting statement explaining why they are suited to this project. To apply, please follow the links at https://www.edinburghneuroscience.ed.ac.uk/project/artificial-intelligence/ai-powered-system-support-neuroradiological-readings-cerebrovascular

Please note, your application may be shared with the funders of this PhD Studentship, Medical Research Scotland and Canon Medical Research Europe Ltd.

Interviews are expected to take place approximately 3 weeks after the closing date for applications. Interviews may be conducted by video conference.

It is anticipated that the PhD Studentship will start 12 September 2022.