NL
Key objectives
- To identify IRD and AMD disease signatures for patient stratification, and to determine the window of opportunity for each intervention modus (gene/pathway/cell-based therapy)
- To define a core set of disease outcomes tailored to the intervention modus for use in clinical trials
Our concept Patient profiling
This platform will build on our former research activities, existing data, and our expertise in clinical ophthalmology, epidemiology, artificial intelligence, prediction modelling, and visual system imaging.
learn moreOur concept Patient profiling
This platform will build on our former research activities, existing data, and our expertise in clinical ophthalmology, epidemiology, artificial intelligence, prediction modelling, and visual system imaging.
The first goal is to uncover disease and protection signatures for IRD and AMD using data-driven approaches to ensure appropriate, effective, and accessible delivery of gene-based, pathway-based, and cell-based interventions for AMD and IRD. The patient cohorts that we will target present a complex temporal and phenotypic heterogeneity, showing distinct disease trajectories with diverse phenotypic variability and temporal progression patterns. Understanding this heterogeneity is key to identify the correct treatment for each patient and the appropriate interval of success. A comprehensive picture of this heterogeneity will be unveiled by automatically disentangling the information on vast amounts of functional and structural data which is currently available to the consortium.
The second goal addresses development of a set of relevant disease outcomes for each pillar to evaluate treatment success; outcomes which are clinically meaningful, validated, reliable, and sensitive to change. We will go beyond visual function, and assess a comprehensive set of imaging and functional tests for IRD and AMD at regular intervals during a ‘trial period’ of 3 years. We will perform temporal structure-function mapping of the retina and relate short term change with the disease signatures discovered within our first goal. We will also evaluate whether cortical responses, patient experience, and mobility in daily life settings should have a place in the design of trial outcomes.
Our approach
Platform B consists of two work packages.
Work Package 5.1: Data-driven disease signatures and patient profiles
Work Package 5.2: Development of a core outcome set for clinical trials
Our innovation
With a data-driven approach which disentangles phenotypic and temporal heterogeneity from structural and functional data, we will have defined the intervention-specific disease signatures.
learn moreOur innovation
The work in this platform is truly innovative in terms of its aims as well as its methodology. With a data-driven approach which disentangles phenotypic and temporal heterogeneity from structural and functional data, we will have defined the intervention-specific disease signatures. The profile from responders and nonresponders to treatment can help validate the model for patient stratification. Hence, we will have generated the evidence to select the right patient for the right treatment at the right time. This will be a huge leap for the design, conduct, and analysis of clinical trials, which in turn will accelerate the approval of effective treatments for retinal disease.
Aside from its goals, platform B also creates a new avenue for AI-based modelling. Using a novel AI-based data analysis engine, discovering new aspects of deep generative models and geometric deep learning, we will coherently analyse and disentangle highly complex data: sparse, spatiotemporal, and multimodal. This will set a new standard for data sciences in ophthalmology and other medical fields, finally achieving a holistic view of the patient for well-informed decision-making.
WP leaders
Clarisa Sánchez
About me
I am a Full Professor of AI and Health at the University of Amsterdam and Amsterdam UMC, dedicated to transforming patient care through responsible, cutting-edge AI technologies. My research focuses on developing, validating, and integrating AI methodologies to tackle medical data challenges across the patient pathway.
Your contribution to Lifelong VISION
Over the past 20 years, my research has advanced machine learning models for analyzing and interpreting clinical data to enhance patient care. In Platform B, I will leverage this expertise to untangle structural and functional multimodal data, identifying disease signatures that enable precise, timely, and personalized treatment for each patient.
Team
Camiel Boon
About me
I am a clinician-scientist, working as an eye doctor and researcher. My research motto is: ‘From bedside to bench, and back’. The highlight of my career thus is without a doubt that I was able to perform some of the first-in-human gene therapy eye surgeries at Amsterdam UMC.
Your contribution to Lifelong VISION
As a doctor, the specific expertise that I bring to Lifelong VISION is a translational view, with an emphasis on novel gene therapy techniques (gene replacement, CRISPR-based techniques) in a multidisciplinary setting. In addition, my aim is to develop better insights into the most sensitive clinical endpoints to assess disease progression and treatment efficacy in patients. I am a team player, enjoying crosstalk and close collaboration between disciplines. This makes Lifelong VISION the perfect stimulating research environment for me and my research group.
Erik Bekkers
About me
I am an associate professor in Geometric Deep Learning at the University of Amsterdam, specializing in symmetry-preserving and structure-aware machine learning methods. My research focuses on developing geometric deep learning techniques for medical imaging and the sciences, leveraging differential geometry and AI to improve data-driven analyses.
Your contribution to Lifelong VISION
I contribute in Platform B my expertise in geometric deep learning, generative modelling, and AI-driven biomarker research, as to enable robust and data-efficient medical image analysis. My work focuses on developing structure-preserving machine learning methods that enhance quantitative imaging, facilitating precision diagnostics and treatment personalization.
Karin van Garderen
About me
Karin van Garderen is a postdoctoral researcher specializing in medical image analysis and AI, working at the department of Ophthalmology, Erasmus MC. Finished her PhD in 2024 at the same institute, department of Radiology and Nuclear Medicine.
Contribution to Lifelong VISION
I have been working at the intersection of medical and technological research since 2018, with the goal to provide insight to medical practitioners using data analysis and AI. In LifeLong vision my task is to bring the multidisciplinary data from different sources together in Platform B, and to develop new methods for patient profiling.
Koen Haak
About me
As Associate Professor in the Department of Cognitive Science and Artificial Intelligence,I lead the Vision and Imaging Data Analytics group. We develop and apply advanced data-analytical approaches to study how eye diseases and brain injuries impact our ability to see in daily life, and how this might be improved.
Contribution to Lifelong VISION
I lead the neuroimaging work-package of Platform B. The aim is to develop novel, MRI-based biomarkers to facilitate stratification and monitoring. By focusing on the visual brain, we aim to bridge between the restoration of visual signals at the level of the retina and the ability of patients to use these signals in daily life.
Magda Smoor
About me
I am an Assistant Professor in the Eye Epidemiology group at Erasmus MC, specializing in genetic and biomolecular research on age-related macular degeneration (AMD). In addition to my research, I manage large national, international, and personal grants.
Your contribution to Lifelong VISION
Within Lifelong VISION, I serve as the link to the RD5000 consortium, for which I manage the national registry and research database. Furthermore, I contribute expertise in AMD, particularly in genetics, complement biology, and lipid metabolism.
Additionally, as a member of the Project Support Office, I co-manage Lifelong VISION, ensuring the project's strategic and operational success.
Yara Lechanteur
About me
I have been working on AMD related research for over 10 years. My current expertise is in diagnostics of the complement system and early-onset age-related macular degeneration.
Your contribution to Lifelong VISION
I am the principal investigator of a large case-control database (>2000 AMD patients and healthy controls) with extensive phenotypic and genetic data. Our work on complement diagnostics at Radboudumc will provide leads for therapy development within the consortium.