Dr. Peter Jaap de Lint

Peter Jaap de Lint studied medicine at the University of Utrecht. After his medicine studies, he did his PhD on ophthalmic physics with Prof. Dr. Dirk van Norren in Utrecht for which he received two scientific prizes: “Dr. P. Binkhorst prize” and “F. C. Thunder Prize”. He specialized in ophthalmology at the Academic Hospital in Utrecht, after which he completed a fellowship for retinal surgery at the Rotterdam Eye Hospital in 2001. His subspecializations are cataract surgery and medical retina. He worked at the Medical Center Alkmaar from 2002 to 2009. He was co-founder, ophthalmologist and medical director of the Noordholland Eye Center from 2009 - 2014. Between 2015 and 2018 he worked as an eye doctor at Zonnestraal Eye Hospital, in Zaandam. He has also been working as an ophthalmologist consultant for Koninklijke Visio since 2010. De Lint regularly gives lectures at (inter) national congresses and is the author and co-author of several publications.

Based on information gathered from the scientific literature, internet, conferences and personal communications, in this talk, a number of practical examples and approaches will be discussed regarding how artificial intelligence (AI) may already support medical decisions in eye care. One of the most successful cases, so far, is screening for diabetic retinopathy based on fundus photography, where several companies, such as IDx and Eyenuk, have AI services available. Other eye diseases which are being investigated for use of AI decision support and/or screening are age-related macular degeneration, retinopathy of prematurity, retinal vein occlusions, cataract and glaucoma. Currently, most of the research approaches make use of data from fundus photography, visual field testing and optical coherence tomography (OCT). In a collaboration between Moorfields’ Eye Hospital in London and Deep Mind (part of Google) it was demonstrated that differential diagnosis based on OCT data could be generated for individual patients which were as accurate as the retina experts were able to produce. Similar results have been found in Vienna in the research group led by Prof. Schmidt-Erfurth. A number of principles will be discussed on how such AI systems work, as well as the so-called “Black Box” principle, the importance of “biomarkers”, unsupervised and supervised learning strategies, and the need for explainable AI. With these developments it seems quite feasible that we will see an increase of eye screening and diagnostic tests taking place outside eye departments in the next ten years.

Peter Jaap’s Linkedin Profile: linkedin.com/in/peter-jaap-de-lint-837525a