Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Accept all cookies Reject all non-essential cookies Find out more

Enhancing automated CT-brain analysis tools for quantifying atrophy and white matter disease and development of a novel CT-brain analysis tool to quantify stroke lesions

Lead supervisor: Professor Sarah Pendlebury

Co-supervisors: Professor Mark Jenkinson, Associate Professor Ludovica Griffanti

Commercial partner: Brainomix

 

 

Background

CT-brain imaging is the standard brain imaging modality used in the NHS and globally and is cheaper and better tolerated than MRI particularly in older, frail patients in whom MRI may be contraindicated. We have shown that white matter disease (WMD) and cerebral atrophy on routinely acquired brain imaging can predict delirium and dementia occurring up to 5-years later (Pendlebury et al, Age Ageing,2021; Pendlebury & Rothwell, Lancet Neurology,2019). CT-brain scans therefore have clinical utility beyond the immediate indication but are not fully exploited because of a lack of automated user-friendly CT-analysis tools. Our team has recently developed a CT-brain analysis tool that quantifies both global and temporal lobe atrophy and WMD in less than 4 seconds (Bal et al, in submission) but external validation and further optimisation and improvements are required. In the proposed project, the student would validate and adapt the tool using external datasets from international collaborators and paired MRI scans. The student will also develop a new CT-brain analysis tool to quantify and locate acute and chronic stroke lesions building on our team’s established expertise in lesion mapping using MRI-brain imaging. The externally validated CT-brain analysis toolset would be transformational in unlocking the data on brain ageing and pathology contained in CT-brain scans with wide application to research and clinical practice.

 

Aim

i)            Externally validate and improve our existing global and regional atrophy and WMD CT-brain analysis tools

ii)           Develop and validate a new CT-brain analysis tool for quantifying stroke lesions.

 

Objectives

•            Refine and optimise our existing CT-atrophy and CT-WMD tools using collaborator CT-brain scans;

•            Evaluate tool performance using reference standards including i) visual ratings of atrophy and WMD ii) automated MRI quantifications from paired MRI  scans where available, iii) manual segmentations iv) benchmarking against existing segmentation tools including from collaborators;

•            Develop a new CT-stroke tool to identify and quantify new/old stroke lesions using adaptations of our existing MRI-based lesion mapping tools (Sundaresan et al. 2021, doi.org/10.48550/arXiv.2105.11356).

 

Our collaborator cohorts include i) a multicentre study in Norway (NorCOAST, n~800 acute stroke, Bayer), ii) Singapore memory clinic with paired CT and MRI scans (n=241, Chen) and iii) Ibadan, Nigeria (n=200, acute stroke, Ogbole).

The CT-stroke tool will be developed using our large, well characterised cohorts of older patients: Oxford and Reading Cognitive Co-morbidity, Frailty and Ageing Research Database-Electronic Patient Records (ORCHARD-EPR n>2000), OCS-Tablet/Recovery (n>850); Oxford Vascular Study (n>1,000). A range of machine learning methods (deep learning and classical methods) will be used and promising new architectures (e.g. transformers). The CT-stroke tool will then be externally validated on collaborator cohorts, as for the other tools, to ensure generalisability.

 

Relevance of the  project to the MRC remit

The project will provide training in quantitative, interdisciplinary research applied to an issue of major importance in healthcare with particular relevance to multimorbidity and dementia prediction, and substantial experience in industrial strategy priority topics including digital technologies, translational development, and precision healthcare and diagnostics. The project will be done within an established multidisciplinary academic team in collaboration with Brainomix, a company with CE-marked CT-brain software for hyperacute stroke decision support (e-ASPECTS) embedded in over 300 hospitals. Brainomix is therefore potentially an ideal partner for future commercialisation via a license deal to which the student’s project will contribute necessary clinical data.  Collaboration with Brainomix will also provide the student with exposure to the commercial environment.

 

 

Apply using course: DPhil in Clinical Neurosciences

MRC logo