MRC Precision Medicine DTP: Virtual biopsy with the Eye – using machine learning to detect and trackchronic kidney disease

Industrial Partner: Optos

Background
CKD affects 6-11% of the world’s population [1]. Hypertension is an independent risk factor for CKD progression and is a frequent finding in patients with CKD. Around a quarter of the world’s adult population is hypertensive, a number that is projected to rise to nearly 30% by 2025 [2]. Renal microvascular changes are important in the development of CKD. Currently, these can only be assessed reliably through renal biopsy, which is not without risk. However, the kidney and eye are remarkably similar in their developmental, structural and pathogenic pathways [3]. Transparency of the ocular media offers a unique opportunity to directly visualize and image microvasculature within the eye that may be affected in systemic diseases such as hypertension and CKD. Optical coherence tomography (OCT) is a non-invasive and rapid method for cross-sectional imaging of the retina and choroid. Our pilot data suggest chorioretinal thinning in CKD and that choroidal thickness associates strongly with measures of kidney injury, systemic and renal inflammation as well as endothelial dysfunction [4]. Retinal imaging metrics may reflect renal (and systemic) injury and could have utility in detection and tracking of kidney disease over time.

Aims
Hypothesis: OCT of the eye will provide a virtual biopsy to aid in the assessment of patients with CKD.
Specifically:
1. Segment the chorioretinal layers and map the choroidal vasculature in 3D by
developing novel post-processing methodologies and machine learning (ML) for OCT.
2. Investigate new choroidal vascular metrics relating to abnormalities in patients with CKD and compare to healthy volunteers (HV) and patients with a similar degree of hypertension but without CKD.
3. Explore whether severity of chorioretinal abnormalities correlates with the degree of CKD and endothelial dysfunction as measured by standard clinical biomarkers.
4. Ascertain whether the retina / choroid is a viable site for biomarkers of disease activity.
The student will be in a unique position to leverage an exciting and rich dataset that has been acquired over the past 2 years from >200 participants. These include HV (~100 subjects), those with hypertension (~100 subjects), those with CKD (~100 subjects) and those with CKD who have received a kidney transplant and therefore, their kidney function has returned to a healthy level (~50 subjects). There may also be the opportunity to collect additional complimentary imaging using state-of-the-art Optos equipment.

Training Outcomes
The student will engage with multiple aspects of eye imaging in patients at increased cardiovascular risk from acquisition to post-processing and analysis. They will also feed into the VAMPIRE project, a joint initiative between the Universities of Edinburgh and Dundee that is building a world-class virtual centre of expertise in retinal biomarkers. The student, with support from members of the VAMPIRE team, will learn about machine learning algorithms (e.g. random forest, neural networks) and apply these as a cutting-edge means of utilising retinal measures to classify the clinical status of study participants. The student will gain key skills by attending appropriate courses (such as those offered by the Education Programme at the Edinburgh CRF on statistical analysis, etc.) and accessing on-line materials offered by the Edinburgh Imaging Academy (e.g. anatomy, physics, biomechanics). They will also interact with the expert members of our group. By the end of the project, the student will be expert in retinal imaging biomarkers of renal and systemic vascular health and will have investigated conventional clinical biomarkers and novel retinal endpoints.

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
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:

http://www.ed.ac.uk/studying/postgraduate/degrees/index.php?r=site/view&id=919

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:

http://www.ed.ac.uk/usher/precision-medicine