Project Details
Projekt
Integrating Imaging, Clinical and Genetic Data with Machine Learning to Establish Biomarkers for Retinal Diseases
Subject Area
Ophthalmology
Medical Informatics and Medical Bioinformatics
Medical Informatics and Medical Bioinformatics
Term
since 2024
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 513025799
High-resolution non-invasive imaging techniques like optical coherence tomography (OCT) are widely used for diagnosing retinal pathologies, especially age-related macular degeneration (AMD). AMD has dry and neovascular forms, with the latter progressing rapidly and causing vision loss. Current treatment involves inhibiting vascular endothelial growth factor (VEGF), but long-term visual acuity stability is limited. The high therapy intensity of anti-VEGF agents poses additional challenges for patients and healthcare systems. Studies show, that from 2009 to 2011 there was an 11-fold increase in annual intravitreal injections, and it is projected to continue to rise in the future. Genetic factors play a significant role in AMD as well, with 52 common variants associated with AMD risk. Some variants seem to influence response to anti-VEGF treatment but their overall impact on disease progression and treatment response is unknown. Current therapeutic guidance lacks a comprehensive utilization of imaging data and integration of genetic information. To address these limitations, artificial intelligence (AI) and machine learning (ML) algorithms, such as convolutional neural networks (CNN), can be employed to process complex medical data and pave the way into better understanding of disease progression and personalized treatment regimen. Therefore, in this project, we aim to incorporate patients’ genetic data from our own longitudinal cohort, leveraging the ever-growing affordability of genetic sequencing. This integration is crucial for common retinal diseases like AMD, which have large imaging datasets suitable for ML. Additionally, with access to cohorts from the UK Biobank, we are uniquely positioned to lead this effort. Our aim is to develop predictive models that integrate imaging, genetics, and clinical data, advancing our understanding of disease progression and therapeutic outcomes. This integrative approach, supported by our computational methods, will create a versatile framework applicable to AMD and other retinal diseases, revolutionizing personalized treatment planning, disease progression assessment, and therapeutic response predictions.
DFG Programme
Research Units
