| Title: | Deep learning and genome-wide analysis to determine the role of reticular pseudodrusen in the progression of age-related macular degeneration |
| Lead Institution: | Moorfields Eye Hospital NHS Foundation Trust |
| Principal investigator: | Mr Adnan Tufail |
Application 60078
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
Brief background and need for the research: Age-related macular degeneration (AMD) is a leading cause of blindness and is characterised in its early stage by deposits under the macula called drusen. Over time, many eyes develop atrophy (worn out patches of the retinal layer that picks up light), also called "dry AMD", for which there is no treatment. A minority of eyes with AMD develop a secondary complication of abnormal blood vessels growth at the back of the eye ("wet AMD") which leads to leakage of fluid into or under the retina. There is currently treatment for wet AMD. It is imperative to better understand the biological pathways and genetic associations that are linked to dry AMD to provide a suitable treatment in the future. Historically, drusen are hallmarks of AMD. However, reticular pseudodrusen (RPD) are another form of abnormal tissue deposits that can mostly be seen in patients with AMD but also with some other conditions that involve the back of the eye. Recently, RPD have been recognised as an important risk factor for advanced AMD, and possibly a greater risk factor than the presence of conventional drusen deposits alone. There is limited understanding of how RPD drives disease. The UK Biobank is an existing large database of patient images and genetic data that have got individual patient consent for research use. Aim of the study: The proposed study addresses important gaps in our knowledge of the key factors for advanced AMD by first developing image recognition software (artificial intelligence tools) to automatically detect RPD. We then intend to apply the software developed in the previous step to detect RPD in existing patient images from the Biobank. The last step will be diagnosing genetic data from the Biobank, to see if previously undiscovered genetic risk factors can be detected.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 10774 | A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography | Roy Schwartz (+15) | 2022 | Translational Vision Science & Technology |
Enabling scientific discoveries that improve human health