A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction

The retinal age gap (RAG; the difference between the retina’s biological and chronological age) has recently gained increased attention as a potential image-based, non-invasive, and accessible biomarker for a broad spectrum of ocular and non-ocular diseases. Traditionally, machine learnin...

Full description

Saved in:
Bibliographic Details
Main Authors: Christopher Nielsen, Matthias Wilms, Nils D. Forkert
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Translational Engineering in Health and Medicine
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11023594/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849417556025147392
author Christopher Nielsen
Matthias Wilms
Nils D. Forkert
author_facet Christopher Nielsen
Matthias Wilms
Nils D. Forkert
author_sort Christopher Nielsen
collection DOAJ
description The retinal age gap (RAG; the difference between the retina’s biological and chronological age) has recently gained increased attention as a potential image-based, non-invasive, and accessible biomarker for a broad spectrum of ocular and non-ocular diseases. Traditionally, machine learning predictions of biological retinal age utilize convolutional neural network (CNN) architectures and data from color fundus photography (CFP). Despite being previously unexplored, the multimodal fusion of two-dimensional CFP with three-dimensional optical coherence tomography (OCT) data has significant potential to enhance retinal age prediction accuracy and the diagnostic utility of the RAG biomarker. Therefore, this work presents a novel foundation model-based framework for multimodal retinal age prediction. Technology or Method: Feature representations from CFP and OCT images were extracted using RETFound, a powerful foundation model for retinal image analysis. These representations were then combined using an innovative fusion strategy to train a lightweight linear regression head model for predicting retinal age. Training and evaluation of the developed multimodal retinal age prediction model was achieved using retinal images from over 80,000 participants in the UK Biobank. Results: The developed multimodal model sets a new benchmark in retinal age prediction (mean absolute error of 2.75 years), outperforming traditional CNN and single-modality approaches. Additionally, multimodal RAG values demonstrated superior performance in classifying patients with diabetes mellitus type 1, multiple sclerosis, and chronic kidney disease, highlighting the clinical relevance of the proposed multimodal approach for non-ocular disease detection. Conclusions: This work demonstrates that multimodal fusion of CFP and OCT significantly improves retinal age prediction and subsequent RAG-based analyses. By leveraging foundation models and multimodal retinal imaging, the proposed approach enhances disease classification accuracy and demonstrates the potential of integrating the RAG into clinical workflows as a scalable, non-invasive screening tool. Significance: The findings underscore the potential of multimodal retinal imaging to transform RAG into a clinically relevant and highly accessible biomarker for disease detection.
format Article
id doaj-art-7974e94c22ef447a96a028df9db2b7d7
institution Kabale University
issn 2168-2372
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Translational Engineering in Health and Medicine
spelling doaj-art-7974e94c22ef447a96a028df9db2b7d72025-08-20T03:32:46ZengIEEEIEEE Journal of Translational Engineering in Health and Medicine2168-23722025-01-011329930910.1109/JTEHM.2025.357659611023594A Novel Foundation Model-Based Framework for Multimodal Retinal Age PredictionChristopher Nielsen0https://orcid.org/0009-0000-5206-4434Matthias Wilms1https://orcid.org/0000-0001-8845-360XNils D. Forkert2https://orcid.org/0000-0003-2556-3224Biomedical Engineering Graduate Program, University of Calgary, Calgary, AB, CanadaDepartment of Radiology, University of Calgary, Calgary, AB, CanadaDepartment of Radiology, University of Calgary, Calgary, AB, CanadaThe retinal age gap (RAG; the difference between the retina’s biological and chronological age) has recently gained increased attention as a potential image-based, non-invasive, and accessible biomarker for a broad spectrum of ocular and non-ocular diseases. Traditionally, machine learning predictions of biological retinal age utilize convolutional neural network (CNN) architectures and data from color fundus photography (CFP). Despite being previously unexplored, the multimodal fusion of two-dimensional CFP with three-dimensional optical coherence tomography (OCT) data has significant potential to enhance retinal age prediction accuracy and the diagnostic utility of the RAG biomarker. Therefore, this work presents a novel foundation model-based framework for multimodal retinal age prediction. Technology or Method: Feature representations from CFP and OCT images were extracted using RETFound, a powerful foundation model for retinal image analysis. These representations were then combined using an innovative fusion strategy to train a lightweight linear regression head model for predicting retinal age. Training and evaluation of the developed multimodal retinal age prediction model was achieved using retinal images from over 80,000 participants in the UK Biobank. Results: The developed multimodal model sets a new benchmark in retinal age prediction (mean absolute error of 2.75 years), outperforming traditional CNN and single-modality approaches. Additionally, multimodal RAG values demonstrated superior performance in classifying patients with diabetes mellitus type 1, multiple sclerosis, and chronic kidney disease, highlighting the clinical relevance of the proposed multimodal approach for non-ocular disease detection. Conclusions: This work demonstrates that multimodal fusion of CFP and OCT significantly improves retinal age prediction and subsequent RAG-based analyses. By leveraging foundation models and multimodal retinal imaging, the proposed approach enhances disease classification accuracy and demonstrates the potential of integrating the RAG into clinical workflows as a scalable, non-invasive screening tool. Significance: The findings underscore the potential of multimodal retinal imaging to transform RAG into a clinically relevant and highly accessible biomarker for disease detection.https://ieeexplore.ieee.org/document/11023594/Foundation modelmachine learningretinal imagingretinal age prediction
spellingShingle Christopher Nielsen
Matthias Wilms
Nils D. Forkert
A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction
IEEE Journal of Translational Engineering in Health and Medicine
Foundation model
machine learning
retinal imaging
retinal age prediction
title A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction
title_full A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction
title_fullStr A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction
title_full_unstemmed A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction
title_short A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction
title_sort novel foundation model based framework for multimodal retinal age prediction
topic Foundation model
machine learning
retinal imaging
retinal age prediction
url https://ieeexplore.ieee.org/document/11023594/
work_keys_str_mv AT christophernielsen anovelfoundationmodelbasedframeworkformultimodalretinalageprediction
AT matthiaswilms anovelfoundationmodelbasedframeworkformultimodalretinalageprediction
AT nilsdforkert anovelfoundationmodelbasedframeworkformultimodalretinalageprediction
AT christophernielsen novelfoundationmodelbasedframeworkformultimodalretinalageprediction
AT matthiaswilms novelfoundationmodelbasedframeworkformultimodalretinalageprediction
AT nilsdforkert novelfoundationmodelbasedframeworkformultimodalretinalageprediction