BACKGROUND: The choroid is a vital vascular layer of the eye, essential for maintaining ocular health. Understanding its structural variations, particularly choroidal thickness (CT), is crucial for the early detection of diseases, such as age-related macular degeneration (AMD), high myopia (HM), and diabetes mellitus (DM). Recent advancements in deep learning have significantly improved the segmentation and measurement of choroidal layers. OBJECTIVE: This study aims to investigate age- and gender-related changes in CT and its components through deep learning analysis of swept-source optical coherence tomography (SS-OCT) images. METHODS: A total of 262 participants (136 females and 126 males) were recruited from Peking University International Hospital. Exclusion criteria included ocular pathologies and systemic conditions. SS-OCT was utilized for CT, Sattler layer-choriocapillaris complex thickness (SLCCT), and Haller layer thickness (HLT) measurements. auto-measurement method, based on deep learning algorithms, ensured accuracy. Ethics approval and informed consent were obtained from all participants. FINDINGS: Significant thinning of CT and SLCCT was observed after the age of 60, with HLT declining after the age of 30. Females exhibited marked thinning between the ages of 40 and 50, while males began to show thinning at age 60. CONCLUSION AND IMPLICATIONS: This research highlights age-related changes in choroidal thickness, with a particular emphasis on gender differences. The findings suggest that females experience earlier thinning, potentially attributable to hormonal changes. Additionally, the study validates the efficiency of deep learning algorithms in measuring choroidal thickness, thereby enhancing the reliability of clinical practice.