关键词: 冠心病, 视网膜血管, 形态学参数, 人工智能

Abstract: Objective To preliminarily analyze the correlation between retinal vascular morphology parameters and 70 years-old or over coronary heart disease using artificial intelligence automated analysis technology. Design Cross-sectional study. Participants 471 70 years-old or over coronary heart disease patients undergoing chronic disease management at Beijing Jiaotong University Hospital from January 2023 to January 2024 were selected as the observation group, and 426 age- and gender-matched individuals who signed up for documentation and were excluded from coronary heart disease during the same period were selected as the control group. Methods General information, fasting blood glucose (FBG), low-density lipoprotein cholesterol (LDL-C), and color fundus photography images of the right eye without dilated pupils were collected from both groups. Retinal vascular morphology parameters were measured using artificial intelligence automated analysis technology to observe the differences between the two groups; binary Logistic regression was used to analyze the correlation between retinal vascular morphology parameters and coronary heart disease. Main Outcome Measures Average caliber of retinal artery, average caliber of retinal vein, average retinal artery tortuosity, average retinal vein tortuosity, retinal vascular branching angle, vascular fractal dimension, and retinal vascular density. Results Compared with the control group, patients in the observation group had a higher proportion of smokers, higher BMI, higher FBG and LDL-C (all P<0.05). Age, gender, proportion of patients with hypertension and diabetes mellitus, and proportion of alcohol consumption showed no significant difference between the two groups. Compared with the control group, patients in the observation group had finer average caliber of retinal artery（t=29.922, P＜0.001）, greater average retinal artery tortuosity（t=-2.174, P=0.03）, lower vascular fractal dimension（t=2.282, P=0.023）, and lower retinal vascular density（t=4.057, P＜0.001）. Binary Logistic regression analysis showed that average caliber of retinal artery was independently associated with coronary artery disease after correction for confounders (B=-0.721, P<0.001). Conclusion Compared with the control group, average caliber of retinal artery was finer and independently associated with coronary artery disease in the observation group of 70 years-old or over patients with coronary artery disease. (Ophthalmol CHN, 2024, 33: 285-289)

Key words: coronary heart disease, retinal vessels, morphological parameters, artificial intelligence