Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Mar;21(3):e14601.
doi: 10.1002/alz.14601.

A deep-learning retinal aging biomarker for cognitive decline and incident dementia

Ming Ann Sim  1   2 , Yih Chung Tham  3   4   5 , Simon Nusinovici  4 , Ten Cheer Quek  4 , Marco Yu  4 , Can Can Xue  4 , Miao Li Chee  4 , Qing Sheng Peng  3 , Eugene S J Tan  6 , Siew Pang Chan  6 , Yuan Cai  7 , Eddie Jun Yi Chong  1 , Boon Yeow Tan  8 , Narayanaswamy Venketasubramanian  9 , Saima Hilal  10 , Mitchell K P Lai  1 , Hyungwon Choi  11 , Arthur Mark Richards  11   12 , Ching-Yu Cheng  3   4   5 , Christopher L H Chen  1
Affiliations

A deep-learning retinal aging biomarker for cognitive decline and incident dementia

Ming Ann Sim et al. Alzheimers Dement. 2025 Mar.

Abstract

Introduction: The utility of retinal photography-derived aging biomarkers for predicting cognitive decline remains under-explored.

Methods: A memory-clinic cohort in Singapore was followed-up for 5 years. RetiPhenoAge, a retinal aging biomarker, was derived from retinal photographs using deep-learning. Using competing risk analysis, we determined the associations of RetiPhenoAge with cognitive decline and dementia, with the UK Biobank utilized as the replication cohort. The associations of RetiPhenoAge with MRI markers(cerebral small vessel disease [CSVD] and neurodegeneration) and its underlying plasma proteomic profile were evaluated.

Results: Of 510 memory-clinic subjects(N = 155 cognitive decline), RetiPhenoAge associated with incident cognitive decline (subdistribution hazard ratio [SHR] 1.34, 95% confidence interval [CI] 1.10-1.64, p = 0.004), and incident dementia (SHR 1.43, 95% CI 1.02-2.01, p = 0.036). In the UK Biobank (N = 33 495), RetiPhenoAge similarly predicted incident dementia (SHR 1.25, 95% CI 1.09-1.41, p = 0.008). RetiPhenoAge significantly associated with CSVD, brain atrophy, and plasma proteomic signatures related to aging.

Discussion: RetiPhenoAge may provide a non-invasive prognostic screening tool for cognitive decline and dementia.

Highlights: RetiPhenoAge, a retinal aging marker, was studied in an Asian memory clinic cohort. Older RetiPhenoAge predicted future cognitive decline and incident dementia. It also linked to neuropathological markers, and plasma proteomic profiles of aging. UK Biobank replication found that RetiPhenoAge predicted 12-year incident dementia. Future studies should validate RetiPhenoAge as a prognostic biomarker for dementia.

Keywords: Southeast‐Asian; biomarker; cognitive decline; deep‐learning; dementia; prognostic; retinal age; retinal photography.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest. Author disclosures are available in the supporting information.

Figures

FIGURE 1
FIGURE 1
Overview of study design (created with biorender.com).
FIGURE 2
FIGURE 2
Flowchart of subject recruitment within the Asian memory clinic cohort and UK Biobank cohort (created with biorender.com). CIND refers to cognitive impairment, no dementia. Cognitive decline was defined as a longitudinal increment in Clinical Dementia Rating scale Sum of Boxes score of 3 points and above, at any annual follow‐up. Q.C. refers to quality control.
FIGURE 3
FIGURE 3
Kaplan–Meier curves of (A) cognitive decline, and RetiPhenoAge stratified by quartiles (Southeast‐Asian memory clinic cohort); (B) conversion to dementia, and RetiPhenoAge stratified by quartiles (Southeast‐Asian memory clinic cohort); (C) incident dementia and RetiPhenoAge stratified by quartiles (UK Biobank cohort). p‐values presented were derived from log‐rank tests of cognitive decline or incident dementia respectively, stratified by RetiPhenoAge quartiles.
FIGURE 4
FIGURE 4
(A) Cross‐sectional associations of RetiPhenoAge with brain MRI markers of cerebral small vessel disease and neurodegeneration. Forest plots depict Exp(β) and adjusted relative risk (RR), with 95% confidence intervals derived from multivariable linear or Poisson regression, respectively. (B) Top 20 most significantly associative plasma proteins for older RetiPhenoAge. Regression coefficients and corrected p‐values (q‐values) were derived from linear regression of plasma proteins and RetiPhenoAge. (C) Top five most significant Gene Ontology (biological process), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Reactome pathways over‐represented within the associative plasma proteins for RetiPhenoAge.
See this image and copyright information in PMC

References

    1. International AsD . World Alzheimer Report 2018, The state of the art of dementia research: New frontiers. 2018;
    1. Hou Y, Dan X, Babbar M, et al. Ageing as a risk factor for neurodegenerative disease. Nat Rev Neurol. 2019;15(10):565‐581. doi: 10.1038/s41582-019-0244-7 - DOI - PubMed
    1. Lopez‐Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194‐1217. doi: 10.1016/j.cell.2013.05.039 - DOI - PMC - PubMed
    1. Hamczyk MR, Nevado RM, Barettino A, Fuster V, Andres V. Biological versus chronological aging: JACC focus seminar. J Am Coll Cardiol. 2020;75(8):919‐930. doi: 10.1016/j.jacc.2019.11.062 - DOI - PubMed
    1. Matt Kaeberlein P. Time for a new strategy in the war on Alzheimer's disease. Public Policy & Aging Report. 2019;29(4):119‐122. doi: 10.1093/ppar/prz020 - DOI