Prediction of Alzheimer's Disease Risk Factors from Retinal Images via Deep Learning: Development and Validation of Biologically Relevant Morphological Associations in the UK Biobank

2026-05-01

AI Summary

This study investigates whether color fundus photographs contain retinal structural features associated with Alzheimer’s disease (AD) risk factors. Leveraging UK Biobank data, the authors employ deep learning models to predict 12 AD-related risk factors and integrate class activation mapping (CAM) with a newly proposed CAM-Score significance metric to systematically identify biologically relevant retinal regions underpinning these predictions. The work provides the first evidence that retinal morphology in fundus images encodes multiple AD risk factors, achieving strong predictive performance (maximum classification AUROC of 0.948 and R² of 0.762 for continuous variables). Significance maps consistently highlight the optic disc and retinal vasculature, with detectable differences already present approximately 8.55 years before clinical AD diagnosis, suggesting a promising avenue for early AD screening.

Abstract

The systemic, metabolic, lifestyle factors have established associations with Alzheimer's Disease (AD) through epidemiologic and AD-specific biomarker studies. Whether colored fundus photography (CFP) contains retinal structural signatures corresponding to these AD-related risk domains remains unclear. To determine whether deep learning (DL) models can predict 12 AD-related risk factors from CFP and to characterize the retinal structures underlying these predictions, thereby assessing whether CFP reflects pathways to AD vulnerability. Using UK Biobank CFPs, DL models were trained using 62,876 images from 44,501 unique participants to predict 12 factors linked to AD incidence: 6 categorical (sex, smoking, sleeplessness, economic status, alcohol use, depression) and 6 continuous (age, age at completing education, BMI, systolic, diastolic blood pressure, HbA1c). Model performance, model saliency, and saliency-derived scores (CAM-Score) were evaluated and compared to retinal morphometry. The scores were also compared between incident-AD cases (average 8.55 years before onset) and matched controls. Performance of DL ranged from AUROC= 0.5654-0.9480 for categorical and R2=-0.0291-0.7620 for continuous factors, outperforming most of the morphometry-machine learning models. Saliency-based score consistently highlighted biologically meaningful regions, particularly the optic nerve head and retinal vasculature. It also aligned with present morphometric variations. Several saliency-based scores differed significantly between incident AD and matched controls, suggesting potential overlap between retinal correlates of risk factors and preclinical AD-associated changes. CFP encodes retinal signatures linked to AD risk factors. Although not diagnostic, DL-derived retinal representations may uncover biologically meaningful risk-related structural changes mirroring the potential AD vulnerability.

Problem

• Alzheimer's Disease
• Retinal Imaging
• Risk Factors
• Deep Learning
• UK Biobank

Innovation

• deep learning
• retinal imaging
• Alzheimer's disease risk factors
• model saliency
• UK Biobank