This study introduces a novel biological aging marker derived from retinal fundus images to assess female reproductive aging. The key measure, termed the retinal age gap, represents the difference between AI-predicted retinal age and chronological age, reflecting how rapidly retinal tissue is aging. Unlike previous retinal aging studies that focused on systemic or neurological diseases, this work is the first to demonstrate that retinal aging specifically mirrors the ovarian aging process.
The scientific novelty lies in showing that the eye functions as a surrogate marker for reproductive aging. Among women aged 45–50, each additional year of retinal age gap was associated with a 20% higher risk of low anti-Müllerian hormone (AMH) levels (OR = 1.20), indicating diminished ovarian reserve. Furthermore, accelerated retinal aging was linked to earlier menopause, with each additional year of retinal age gap increasing the risk of menopause before age 45 by 22% to 36%. These associations remained significant after adjustment for chronological age, demonstrating that retinal aging captures reproductive risk beyond age alone.
Methodologically, the study introduces a dual-channel AI framework that preserves general aging patterns while learning female-specific reproductive aging signals. This design enables the model to detect subtle vascular and hormonal changes that standard aging models often miss. In addition, genome-wide analyses revealed that retinal aging and reproductive senescence share common biological pathways, including those involved in progesterone-mediated oocyte maturation and oxytocin signaling, providing genetic evidence for a shared eye–ovary aging mechanism.
By further integrating retinal images with genetic markers, the study demonstrates that hormone-related aging risk can be inferred without direct blood testing. The impact of this work is the establishment of a non-invasive, scalable approach for early identification of women at risk of accelerated reproductive aging, supporting proactive fertility counseling and personalized reproductive health management.
Miao, H., Liu, S., Wang, Z., Zhang, X., Li, Y., & Chen, J. (2025). Artificial intelligence-derived retinal age gap as a marker for reproductive aging in women. npj Digital Medicine, 8, Article 367. https://doi.org/10.1038/s41746-025-01699-8
