AI-based oculomics for trajectory-driven risk stratification of pathologic myopia in paediatric high myopia - PubMed
5 hours ago
- #Paediatric High Myopia
- #Risk Stratification
- #AI Oculomics
- AI-based oculomics model developed for early risk stratification of pathologic myopia (PM) in children with high myopia (HM).
- Longitudinal study of 375 children with HM followed for ~15 years, using deep learning on retinal images and ocular biometry to identify PM predictors.
- Key predictors include lower retinal vessel density, reduced fractal dimension, narrower arteriolar/venular calibres, thinner subfoveal choroid, and faster early axial elongation/choroidal thinning.
- AI model achieved excellent discrimination (AUROC 0.96-0.98), integrated into a web-based prediction tool named 'System for Myopia AI-based Risk Tracking-PM'.
- Study suggests AI-enabled oculomic profiling supports targeted surveillance and prevention of sight-threatening PM in paediatric HM.