Background <p>The Atherogenic Index of Plasma (AIP) is a novel logarithmic index that combines fasting triglyceride and high-density lipoprotein cholesterol (HDL-C) concentrations and is associated with the burden of atherosclerosis. Currently, the noninvasive nature of fundus photography combined with the rapid development of deep learning techniques for quantitative retinal microvascular parameters is advancing rapidly. The aim of this study was to evaluate the correlation between retinal vascular characteristics and AIP using quantitative analysis of retinal vessels in the fundus, thereby further exploring the relationship between glucose and lipid metabolism and atherosclerotic markers with microvascular sclerosis.</p> Methods <p>This study was based on a cohort of 651 inpatients from the Cardiovascular Center database of Beijing Tongren Hospital, Capital Medical University. Patients were divided into four groups (Q1-Q4) according to the quartiles of the AIP index. Fundus images were captured using a nonmydriatic 45° fundus camera for each patient. A deep learning–based segmentation model was employed to perform fully automated quantitative measurements of retinal vascular parameters on color fundus photographs (CFP), yielding multiple quantitative fundus data points, including fractal dimension (FD), vessel density (VD), and mean blood vessel diameter (MBVD). Linear regression and subgroup analyses were conducted to investigate the correlation between retinal vascular characteristics and AIP.</p> Results <p>In this study, significant differences were observed among the groups in terms of age, history of hyperlipidemia, diabetes, smoking, number of patients diagnosed with coronary artery disease (CAD) during the current admission, Gensini score from coronary angiography during the current admission, BMI, fasting blood glucose (FBG), total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), HDL-C, and glycated hemoglobin (HbA1c) (<i>P</i> &lt; 0.05). In the quantitative retinal vascular results, significant differences were found among the groups in fractal dimension (FD), vessel density (VD), mean blood vessel diameter (MBVD), and mean venous diameter (MVD) (<i>P</i> &lt; 0.05). Further comparisons revealed that FD differed significantly between Groups Q1 and Q4, as well as between Groups Q2 and Q4, while VD, MBVD, and MVD differed significantly between Groups Q1 and Q4 (<i>P</i> &lt; 0.05). In the linear regression analysis between AIP levels and ocular quantitative parameters, after adjusting for age, sex, medical history, personal history, BMI, and results of glucose and lipid tests (Model 4), FD and VD were negatively correlated with AIP levels. In addition, the restricted cubic spline (RCS) analysis indicated no significant nonlinear relationship between AIP and FD or VD (<i>P</i> for nonlinear &gt; 0.05). In subgroup analyses, the associations between AIP and FD as well as VD were more pronounced in individuals aged ≥ 65&#xa0;years (<i>P</i> for interaction &lt; 0.05).</p> Conclusion <p>AIP levels are significantly negatively correlated with FD and VD in ocular quantitative results. Subgroup analyses revealed that in individuals aged ≥ 65&#xa0;years, FD and VD are correlated with AIP levels, with significant interaction effects.</p>