Retinal age gap as a predictive biomarker of future risk of Parkinson’s disease

• DOI: 10.1093/ageing/afac062
• Publication Date: 2022-02-21T12:09:22Z
• Authors: Wenyi Hu; Wei Wang; Yueye Wang; Yifan Chen; Xianwen Shang; Huan Liao; Yu Huang; Gabriella Bulloch; Shiran Zhang; Katerina Kiburg; Xueli Zhang; Shulin Tang; Honghua Yu; Xiaohong Yang; Mingguang He; Zhuoting Zhu
• Keywords: Quartile; Fundus (uterus)

retinal age derived from fundus images using deep learning has been verified as a novel biomarker of ageing. We aim to investigate the association between gap (retinal age-chronological age) and incident Parkinson's disease (PD).a (DL) model trained on 19,200 11,052 chronic disease-free participants was used predict age. Retinal generated by DL for remaining 35,834 free PD at baseline assessment. Cox proportional hazards regression models were utilised PD. Multivariable logistic applied prediction 5-year risk area under receiver operator characteristic curves (AUC) estimate predictive value.a total (56.7 ± 8.04 years, 55.7% female) included in present analysis. After adjustment confounding factors, 1-year increase associated with 10% (hazard ratio [HR] = 1.10, 95% confidence interval [CI]: 1.01-1.20, P 0.023). Compared lowest quartile gap, significantly increased third fourth quartiles (HR 2.66, CI: 1.13-6.22, 0.024; HR 4.86, 1.59-14.8, 0.005, respectively). The value established factors comparable (AUC 0.708 0.717, 0.821).retinal demonstrated potential identifying individuals high developing future