Deep Imbalanced Regression Model for Predicting Refractive Error from Retinal Photos
Recent studies utilized ocular images and deep learning (DL) to predict refractive error and yielded notable results. However, most studies did not address biases from imbalanced datasets or conduct external validations. To address these gaps, this study aimed to integrate Deep Imbalanced Regression (DIR) technique into ResNet and Vision Transformer models to predict refractive error from retinal photographs.