Purpose: Glaucoma is a chronic eye disease that progressively damages the optic nerve, leading to irreversible visual field loss.Optical Coherence Tomography (OCT) and Visual Field (VF) tests are essential for monitoring structural and functional changes in glaucoma.This study applies three deep learning models-R2 U-Net, Dense U-Net, and Nested U-Net (UNet++)-to predict visual field outcomes using Retinal Nerve Fiber Layer (RNFL) thickness maps from OCT images. Design:Retrospective cross-sectional study.Subjects: A total of 1,640 patients with glaucoma diagnosed at a tertiary referral center were included.Only one eye (left eye) per patient was analyzed to avoid inter-eye correlation.Eyes included patients with early, moderate, and advanced glaucoma.Methods: We used a dataset of OCT and VF data from 1640 glaucoma patients, divided into training, validation, and test sets.The three deep learning models were trained and evaluated using five performance metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE), Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Frchet Inception Distance (FID).The goal was to predict visual field outcomes based on OCT-derived RNFL thickness maps.Main Outcome measures: Accuracy and image quality of predicted VF maps compared with ground truth VF maps, assessed by MSE, MAE, PSNR, SSIM, and FID.Results: R2 U-Net outperformed all other architectures, with the lowest MSE and MAE values and the highest SSIM and PSNR scores, indicating superior accuracy and image quality.Nested U-Net and Dense U-Net lagged, with Dense U-Net showing the lowest predictive accuracy.J o u r n a l P r e -p r o o f Conclusion: This study is the first to apply generative AI models, such as R2 U-Net, to predict visual field loss based on OCT data, with both models demonstrating exceptional performance.These findings highlight the potential of generative AI to enhance glaucoma diagnosis and facilitate personalized treatment planning.
Ohn et al. (2026) studied this question.