AUTHOR=Chen Siyu , Mu Jingyu , Tan Xiyue , Wu Xiaoxiao , Duan Junguo 

TITLE=Prediction of myopia based on biometric parameters of 500,000 children and adolescents aged 3–18 years

JOURNAL=Frontiers in Public Health

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2025.1563305

DOI=10.3389/fpubh.2025.1563305

ISSN=2296-2565

ABSTRACT=PurposeTo explore the validity and significance of the axial length/corneal radius (AL/CR) Ratio for myopia prediction in children of all ages.MethodsBetween 2020 and 2022, 509,530 children and adolescents aged 3–18 years were sampled in Chengdu City, Sichuan Province, China, by whole-cluster sampling method. Measured their uncorrected visual acuity (UCVA), non-cycloplegic autorefraction, axial length (AL), and corneal radius (CR), Pearson's correlation coefficient and Receiver operating characteristic (ROC) curve were used to determine the accuracy and calculate the cutoff for myopia detection.ResultsThe correlation between AL/CR ratio and SE was higher than that between AL and SE at different ages and refractive states. The area under the ROC curve (AUC) of myopia detection by AL/CR ratio (0.9112) was significantly larger than that of AL (0.8923, P < 0.001), The difference between boys and girls using AL/CR ratio to detect myopia was also statistically significant (p < 0.001). The cutoffs for predicting myopia by AL/CR ratio increased with age, from >2.755 in 3-year-old to >3.095 in 18-year-old, with boys increasing from >2.755 to >3.095, and girls from >2.715 to >3.085, and the cutoff for boys was higher than that for girls at the same age.ConclusionDifferent cutoff for AL/CR ratio can be used to predict myopia for children and adolescents of different ages and genders, and this method can be widely used for clinical diagnosis and mass myopia screening.