Developing new diabetic retinopathy biomarkers through image processing, computational modelling, and machine learning.
Lead Research Organisation:
University of Edinburgh
Department Name: Sch of Molecular. Genetics & Pop Health
Abstract
The aim of this project is to investigate clinically relevant diabetic retinopathy biomarkers and design the computational pipelines and protocols necessary to facilitate their use in future large-scale clinical and pre-clinical studies.
Diabetic retinopathy (DR) is the leading cause of visual loss in developed countries worldwide. Previous studies have reported structural and haemodynamic changes in the diabetic eye microvasculature that precede clinically manifested pathological alterations of the retina. Therefore, new methods that allow greater understanding of these early changes may empower earlier detection of DR helping to limit damage to the retina and
vision function.
Optical Coherence Tomography Angiography (OCT-A) is a non-invasive, dyeless technique capable of resolving the microvasculature of the eye to a level of detail never
seen before. It is only very recently that the first OCT-A commercial products have started to be utilised in clinical centres. One of the first OCT-A devices installed in the UK is hosted in Edinburgh at the Clinical Research Imaging Centre (CRIC, a clinical research facility in partnership between the University of Edinburgh and NHS Lothian).
Recent works showed that it is possible to build computational blood flow models from high-resolution images of the parafoveal region of the retina (of paramount importance for sharp central vision and visual detail). These computational models can provide a detailed haemodynamic characterisation of the region beyond the capabilities of current imaging technologies.
The project involves design and run a series of longitudinal studies aimed at identifying morphological and flow-based early indicators of the pathological changes that lead to visual loss in advanced DR stages. In particular, image processing, computational modelling, and machine learning methods will be exploited in order to characterise the early changes in microvascular haemodynamics. Moreover, there is a unique opportunity to look at the genetic correlation between omics data and imaging data. Therefore, the project aims to combine these novel methodologies with the OCT-A datasets at Edinburgh and Belfast in order to realise the full potential of this patient-specific approach to achieving Precision Medicine for eye care.
Diabetic retinopathy (DR) is the leading cause of visual loss in developed countries worldwide. Previous studies have reported structural and haemodynamic changes in the diabetic eye microvasculature that precede clinically manifested pathological alterations of the retina. Therefore, new methods that allow greater understanding of these early changes may empower earlier detection of DR helping to limit damage to the retina and
vision function.
Optical Coherence Tomography Angiography (OCT-A) is a non-invasive, dyeless technique capable of resolving the microvasculature of the eye to a level of detail never
seen before. It is only very recently that the first OCT-A commercial products have started to be utilised in clinical centres. One of the first OCT-A devices installed in the UK is hosted in Edinburgh at the Clinical Research Imaging Centre (CRIC, a clinical research facility in partnership between the University of Edinburgh and NHS Lothian).
Recent works showed that it is possible to build computational blood flow models from high-resolution images of the parafoveal region of the retina (of paramount importance for sharp central vision and visual detail). These computational models can provide a detailed haemodynamic characterisation of the region beyond the capabilities of current imaging technologies.
The project involves design and run a series of longitudinal studies aimed at identifying morphological and flow-based early indicators of the pathological changes that lead to visual loss in advanced DR stages. In particular, image processing, computational modelling, and machine learning methods will be exploited in order to characterise the early changes in microvascular haemodynamics. Moreover, there is a unique opportunity to look at the genetic correlation between omics data and imaging data. Therefore, the project aims to combine these novel methodologies with the OCT-A datasets at Edinburgh and Belfast in order to realise the full potential of this patient-specific approach to achieving Precision Medicine for eye care.
Organisations
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
MR/N013166/1 | 01/10/2016 | 30/09/2025 | |||
1938308 | Studentship | MR/N013166/1 | 01/09/2017 | 31/05/2021 | Ylenia Giarratano |
Title | Optical Coherence Tomography Angiography retinal scans and segmentations |
Description | Optical Coherence Tomography Angiography retinal scans from 11 participants in the PREVENT study and associated manual segmentations of the vasculature in the scans. Optical coherence tomography angiography (OCTA) is a novel non-invasive imaging modality for the visualisation of microvasculature in vivo. OCTA has encountered broad adoption in retinal research. OCTA potential in the assessment of pathological conditions and the reproducibility of studies relies on the quality of the image analysis. However, automated segmentation of parafoveal OCTA images is still an open problem in the field. In this study, we generate the first open dataset of retinal parafoveal OCTA images with associated ground truth manual segmentations. Imaging was performed using the commercial RTVue-XR Avanti OCT system (OptoVue, Fremont, CA). Consequent B-scans, each one consisting of 304×304 A-scans, were generated in 3×3 mm field of view centered at the fovea. In this work, we selected images only of the superficial layer (containing the vasculature enclosed in the internal limiting membrane layer (ILM) and the inner plexiform layer (IPL)) from left and right eyes of 11 participants with and without family history of dementia as part of a prospective study aimed to find early biomarkers of neurodegenerative diseases (PREVENT). For each of those images we extracted five subimages, one from each clinical region of interest (ROI): superior, nasal, inferior, temporal, and fovea. Poor quality ROIs were discarded and from the remaining a dataset containing 55 ROIs was created. |
Type Of Material | Database/Collection of data |
Year Produced | 2019 |
Provided To Others? | Yes |
Impact | This dataset was used in study, currently under review, to assess the best methodology to process Optical Coherence Tomography Angiography images |
URL | https://datashare.is.ed.ac.uk/handle/10283/3528 |
Title | OCTA segmentation public code |
Description | Open access code for Optical coherence tomography angiography (OCTA) image segmentation, associated to the dataset in https://doi.org/10.7488/ds/2729, and article current under review http://arxiv.org/abs/1912.09978. |
Type Of Technology | Software |
Year Produced | 2019 |
Open Source License? | Yes |
Impact | This software is currently used in my research project as OCTA image preprocessing tool. |
URL | https://github.com/giaylenia/OCTA_segm_study |
Description | Edinburgh Science Festival |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Schools |
Results and Impact | The workshop provided a hands-on, playful introduction to computer programming for kids aged 8+. During the session we discussed how researchers use coding in their daily life, reporting our own experiences. |
Year(s) Of Engagement Activity | 2019 |
URL | https://www.sciencefestival.co.uk/event-details/game-on |
Description | Explorathon 2019 |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Public/other audiences |
Results and Impact | The aim of the event was to provide a chance for general audience to meet researches and get creative with hands-on activities based on cutting-edge research in a friendly environment. |
Year(s) Of Engagement Activity | 2019 |
URL | https://www.explorathon.co.uk/edinburgh/ |
Description | Social Event Officer for the MICCAI Student Board |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Postgraduate students |
Results and Impact | I organised webinars as part of a new initiative to support for students and researchers in the MICCAI community. The first webinar was about Mental Health during the covid-19 pandemic (July 2020). Other two webinars were organised during the MICCAI2020 online conference. |
Year(s) Of Engagement Activity | 2020 |
URL | https://twitter.com/MiccaiStudents/status/1280155658839523334?s=20 |