Artificial intelligence (AI) was highlighted during the
Note: Transcript edited for clarity and length.
Priv.-Doz. Dr. med. Jan H. Terheyden, FEBO:The biggest impact comes from AI applied to retinal imaging, particularly for diabetic retinopathy and other retinal diseases, where algorithms approach or match expert performance and can be deployed at scale. From a research perspective, these applications are successful because they leverage large, standardized imaging data sets and address obvious clinical bottlenecks in screening and triage. At the same time, AI tools that support medical workflows, such as ambient clinical scribes and automated documentation tools, are beginning to have a significant impact by reducing administrative burden and allowing ophthalmologists to focus more on patient care.
Terheyden: A key challenge is translating strong algorithmic performance into real clinical value, integrating AI seamlessly into workflows without adding cognitive load or time. From my perspective, issues of generalizability, explainability, and confidence are central, especially when models are trained on curated datasets that may not reflect real-world populations or imaging variability.
Terheyden: AI can improve efficiency by automating repetitive tasks such as image quality control, segmentation and risk stratification, allowing clinicians to focus on complex decision-making. I see particularly promising potential in combining longitudinal data and multimodal inputs, such as imaging with clinical or functional data, to improve early detection and assessment of disease progression, particularly in glaucoma and retinal diseases.
Terheyden: I believe the greatest advances will come from clinically integrated, multi-modal and forward-looking AI systems, including decision support tools, rather than standalone diagnostic tools. From a research perspective, advances in core models, real-world validation, and regulatory-ready evaluation frameworks designed to perform at the speed at which AI tools are developed will be critical to moving AI from proof of concept to routine and equitable eye care.
Priv.-Doz. Dr. med. Jan H. Terheyden, FEBO
E : [email protected]
Facharzt for Augenheilkunde | Ophthalmologist
Universitäts-Augenklinik Bonn
Department of Ophthalmology, University of Bonn
Bonn, Germany
Disclosures:
Consultant at Bayer
Research funding from Bayer, Roche, Novartis
Fees from Bayer, Novartis, Okko
Reference
-
Terheyden JH. How artificial intelligence supports ophthalmologists today. Presented at: 2nd International Glaucoma Symposium; January 31, 2026; Mainz, Germany.
