A retrospective analysis of patients from Shanghai Parkway Health Ophthalmology Department and Peking University People's Hospital Optometry Center. The dataset likely contains demographic and ocular biometric parameters used to develop a machine learning model for predicting myopia progression. The model was developed by Peng Zhou and published on figshare in April 2026.
Use Cases
- Building machine learning models to predict myopia progression risk based on biometric parameters.
- Analyzing the importance of specific ocular features like pupil size and corneal thickness for axial elongation.
- Comparing the performance of different ML algorithms (e.g., XGBoost, random forest) on clinical prediction tasks.
- Creating quantitative assessment tools for early clinical intervention in myopia management.
Strengths
- The dataset is associated with a published study where an XGBoost model achieved an R² of 0.913 on the training set.
- Variable importance analysis identified specific predictors, with pupil size scoring 100 and corneal thickness scoring 40.88.
- The training and validation sets showed no statistically significant differences in baseline characteristics (all P > 0.05).
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- The dataset size is 9.6 KB, indicating a very limited scope.
Provenance
- Source
- Peng Zhou
- Collection Method
- Retrospective analysis of patient records.
- Freshness
- Last updated 2026-04-24 17:43:41; freshness should be verified.
- Geography
- Patients treated in Shanghai and Beijing, China.