Leveraging AI and in Silico Modeling in Uveitis

Marion MunkSwitzerland Speaker Leveraging AI and in silico modeling in uveitsArtificial intelligence (AI) and in silico modeling hold growing potential in improving our understanding and clinical management of uveitis. This work highlights two complementary approaches: the use of AI to identify imaging-based risk factors for disease progression, and the application of computational biology to investigate potential immunological mechanisms such as molecular mimicry. Based on longitudinal clinical imaging data, machine learning tools were applied to extract and analyze relevant biomarkers with the aim of predicting inflammatory complications. In parallel, bioinformatic methods were used to explore structural and functional similarities between microbial and ocular proteins, supporting hypotheses around immune-mediated tissue damage. Together, these approaches demonstrate how AI-driven analysis and in silico tools can contribute to both individualized disease monitoring and a deeper insight into uveitis pathophysiologyLeveraging Bioinformatics to Identify Targetable Mechanisms in Diabetic Retinal DiseaseThis presentation highlights a bioinformatics-driven approach to understanding how different retinal cells respond to diabetic conditions, with the goal of identifying novel pathways relevant to disease progression and potential therapeutic intervention. By analyzing large-scale transcriptomic datasets from retinal tissue, gene expression changes specific to retinal cells can be mapped to key metabolic and inflammatory signaling networks. This method enables the discovery of altered pathways that may not be apparent through conventional analysis, providing deeper insight into the cellular mechanisms driving diabetic retinopathy. Focusing on pathway-level changes—such as those related to lipid metabolism, cytokine signaling, and cellular stress—this approach offers a powerful tool to uncover molecular targets that could be leveraged for future drug development. The integration of computational biology with retinal cell-specific data opens new avenues for precision medicine and the development of targeted therapies in diabetic retinal disease.