Advances in ocular aging: combining deep learning, imaging, and liquid biopsy biomarkers

Abstract

Ageing is a significant risk factor for a wide range of human diseases. Yet, its direct relationship with ocular ageing as a marker for overall age-related diseases and mortality still needs to be explored. Non-invasive and minimally invasive methods, including biomarkers detected through ocular imaging or liquid biopsies from the aqueous humour or vitreous body, provide a promising avenue for assessing ocular ageing. These approaches are particularly valuable given the eye's limited regenerative capacity, where tissue damage can result in irreversible harm. In recent years, artificial intelligence (AI), particularly deep learning, has revolutionized medical research, offering novel perspectives on the ageing process. This review highlights how integrating deep learning with advanced imaging and liquid biopsy biomarkers has become a transformative approach to understanding ocular ageing and its implications for systemic health.

Keywords: age-related eye diseases; deep learning; imaging; liquid biopsy; ocular aging.

FIGURE 1

Retinal imaging techniques such as OCT (optical coherence tomography), OCTA (optical coherence tomography angiography), and RNFL (retinal nerve fiber layer) are combined with deep learning algorithms to analyze eye conditions. The process involves identifying cataracts, pterygium, and arcus senilis, and calculating a Reti-aging score using multimodal analysis. Additionally, it aids in screening and identifying conditions like age-related macular degeneration (AMD), Parkinson’s disease (PD), and cognitive impairment (CI).

FIGURE 2

This diagram outlines a process for analyzing proteins in eye fluids to determine eye age. It begins with the extraction of proteins from tears, aqueous humor, and vitreous humor. These proteins are then separated and analyzed using proteomics, which involves a simple protein system (open). The identified proteins are associated with various immune cells, such as T cells, B cells, and Treg cells, as well as retinal and retinal nerve cells. The data is combined with artificial intelligence (AI) to assess eye age, potentially offering insights into ocular health and aging.