AI meets Imaging: Predicting Keratoconus Progression with Multimodal Data

Description

Keratoconus is an ocular disease primarily affecting young individuals, characterized by progressive, irreversible corneal thinning. Late diagnosis can lead to severe visual impairment, potentially progressing to blindness. Early detection and regular monitoring are crucial for timely interventions like corneal cross-linking (CXL) to stabilize progression and preserve vision. High-resolution anterior segment optical coherence tomography (OCT) is the gold standard, capturing 25 radial scans with 31,232 measuring points on the anterior ocular surface. This study employs a multimodal AI-based predictive model, integrating high-resolution imaging with clinical indicators. Additionally, socio-economic factors such as education, residential location, and access to specialized treatment centers are analyzed for their impact on diagnosis and monitoring. The goal is to develop an AI model (symbolic and subsymbolic AI) that predicts keratoconus progression and identifies high-Risk patients for early CXL, enabling a personalized approach to disease management.

Period	10 Apr 2025
Event title	Forschungsinteraktion TNF - MED 2025
Event type	Conference
Location	Linz, AustriaShow on map
Degree of Recognition	Local