Relationship between Brachial-Ankle Pulse Wave Velocity and Fundus Arteriolar Area Calculated Using a Deep-Learning Algorithm.

PURPOSE:Retinal vessels reflect alterations related to hypertension and arteriosclerosis in the physical status. Previously, we had reported a deep-learning algorithm for automatically detecting retinal vessels and measuring the total retinal vascular area in fundus photographs (VAFP). Herein, we investigated the relationship between VAFP and brachial-ankle pulse wave velocity (baPWV), which is the gold standard for arterial stiffness assessment in clinical practice. METHODS:Retinal photographs (n = 696) obtained from 372 individuals who visited the Keijinkai Maruyama Clinic for regular health checkups were used to analyze VAFP. Additionally, the baPWV was measured for each patient. Automatic retinal-vessel segmentation was performed using our deep-learning algorithm, and the total arteriolar area (AA) and total venular area (VA) were measured. Correlations between baPWV and several parameters, including AA and VA, were assessed. RESULTS:The baPWV was negatively correlated with AA (R = -0.40, n = 696, P < 2.2e-16) and VA (R = -0.36, n = 696, P < 2.2e-16). Independent variables (AA, sex, age, and systolic blood pressure) selected using the stepwise method showed a significant correlation with baPWV. The estimated baPWV, calculated using a regression equation with variables including AA, showed a better correlation with the measured baPWV (R = 0.70, n = 696, P < 2.2e-16) than the estimated value without AA (R = 0.68, n = 696, P < 2.2e-16). CONCLUSIONS:AA and VA were significantly correlated with baPWV. Moreover, baPWV estimated using AA correlated well with the actual baPWV. VAFP may serve as an alternative biomarker for evaluating systemic arterial stiffness.