Artificial intelligence-quantified fundus tessellated density and its association with myopia severity in Chinese children: a cross-sectional study.

Introduction: To quantitatively assess the whole fundus tessellated density (FTD) using artificial intelligence (AI) and evaluate its association with myopia severity in Chinese children with myopia. Methods: A cross-sectional analysis was conducted on participants with myopia at Beijing Children's Hospital Capital Medical University, from October 2023 to June 2024. AI technology was used to quantitatively measure FTD based on fundus photographs. Participants were classified into mild, moderate, and high myopia (HM) groups based on cycloplegic spherical equivalent (SE). To explore the relationship between FTD and SE, Spearman correlation coefficients along with linear regression analysis were utilized. The ability of FTD to discriminate HM was assessed using Receiver Operating Characteristic curve. Results: The study included 263 eyes from children aged 6-12 years. AI-quantified FTD showed excellent agreement with manual grading of fundus tessellation (area under the curve [AUC] = 0.97, 95% confidence interval [CI]: 0.95-0.98). Median FTD values were 0.01 (0.00-0.02), 0.01 (0.00-0.03), and 0.04 (0.01-0.12) for the three groups respectively. Significant difference was found when comparing HM group with the mild and moderate myopia groups (both Conclusion: AI -quantified FTD demonstrates a significant correlation with the severity of myopia.