About the Speaker
Hrvoje Bogunović studied computer science at the University of Zagreb (Croatia) and completed his PhD in artificial intelligence (AI) for medical imaging at Pompeu Fabra University (Barcelona, Spain). He subsequently joined the University of Iowa (Iowa City, US) as a postdoc, where he started to specialize in AI methodologies with applications in retinal image analysis. In 2015, he joined the Medical University of Vienna, where in an interdisciplinary environment of the Department of Ophthalmology he advanced AI methods for retinal optical coherence tomography (OCT). Since 2021 he has been leading the Christian Doppler Lab for Artificial Intelligence in Retina and in 2024 he took up a tenure-track Assistant Professorship for Medical Image Computing at the AI Institute, Center for Medical Data Science.
He has published over 150 journal and conference articles, serves on the editorial board of IEEE Transactions on Medical Imaging, and Medical Image Analysis, and was recently awarded an ERC Consolidator grant (2024). His research interest focus on advancing the state of the art of AI methods for medical imaging and computer vision, with an emphasis on disease progression modeling and applications in ophthalmology and retinal imaging.
Talk Abstract
Ophthalmology and the eye are at the forefront of artificial intelligence (AI) applications in medicine, including the first fully autonomous FDA-approved diagnostic system in all of medicine. This talk explores how retinal imaging, paired with modern machine learning, can deliver accurate, scalable, and cost-effective decision support for screening, diagnosis, prognosis, and disease monitoring in everyday care. I will highlight recent advances across imaging types (fundus photography and optical coherence tomography), and go beyond retinal diseases using the paradigm of “eye as a window to the body”. Finally, I will discuss how self-supervised and foundation-model approaches are improving performance and robustness in real-world settings, and outline workflows for deployment in screening programs.
We are looking forward to the talk!