01 June 2023

A person's aging can be tracked with a retinal scan

Researchers have proposed a new method for assessing the trajectory of aging: based on AI analysis of retinal changes.

Researchers from Buck Institute for Aging Research, Google Health and Zuckerberg Hospital in San Francisco have developed a technology to assess human aging based on analysis of eye fundus images. The eyeAge method makes it possible to assess even short-term changes and is suitable for evaluating the effectiveness of measures aimed at slowing down age-related changes.

Minor changes in retinal capillaries often go unnoticed, but they can be identified with the help of a trained neural network, the scientists explain. Google researchers previously developed models to predict diabetic retinopathy from retinal images and used them to identify at least 39 eye diseases.
To estimate age, the researchers trained this model on a dataset that included dynamic observations for more than 100,000 patients. The finished model was tested on retinal images of 64,000 people obtained from a British biobank. 

The study showed a high correlation between age estimated using eyeAge and chronological age. In doing so, the results show a more accurate positive prediction coefficient for two consecutive visits to individuals, rather than for random, age-matched people.

The results show that potentially in less than a year we can determine the trajectory of aging with 71% accuracy, noting noticeable changes in the eyes of those undergoing treatment, providing an effective assessment of geroprotective therapy.

The researchers note that the method currently used to assess aging is phenotypic age assessment, which is based on the analysis of biomarkers from the blood. This method is widely accepted, but it is poorly suited for assessing short-term changes. In contrast, the thin blood vessels in the retina record all changes in the body and are independent of food intake or infectious disease. Together with other tools, they can be used to assess the effects of therapies aimed at slowing down aging.

Source: Longitudinal fundus imaging and its genome-wide association analysis provide evidence for a human retinal aging clock | eLife (elifesciences.org)

Found a typo? Select it and press ctrl + enter Print version

Related posts