Journal: Optometry and Vision Science
Article Title: Association of retinal image–based, deep learning cardiac BioAge with telomere length and cardiovascular biomarkers
doi: 10.1097/OPX.0000000000002158
Figure Lengend Snippet: Diagrammatic representation of how the deep learning model and aggregation models are integrated to produce an individual's cardiac BioAge. *CLAiR is the deep learning model that produces the baseline cardiovascular risk score and comprises a series of deep learning models that first identifies which eye is being assessed, before analyzing the retina for a variety of biomarkers as described. *K-NN = K-nearest neighbors method. BioAge = biological age.
Article Snippet: Multiple confounding risk factors and reverse causation mean that any observed associations between leukocyte telomere length and cardiovascular disease should be interpreted with caution; however, studies using Mendelian randomization provide compelling evidence for a relationship between leukocyte telomere length shortening and an increased risk of both atherosclerotic cardiovascular events and an increased risk of hypertension., , Irrespective of the mechanism by which leukocyte telomeres shorten, our findings indicate that, whether assessed by traditional biomarkers or leukocyte telomere length, our novel cardiac BioAge DL model could be used as a novel surrogate biomarker for cardiac health.
Techniques: