Development and application of an artificial intelligence-based quantitative system for white without pressure in myopic retinopathy

White Without Pressure (WWOP) is a common but under-recognized peripheral retinal finding associated with myopia-related complications, yet its objective assessment remains challenging. This study aimed to develop and validate an artificial intelligence (AI) system for the automated segmentation and quantitative assessment of WWOP lesions from ultra-widefield color fundus photographs. The system employs deep learning algorithms for automatic identification and multi-dimensional quantification of WWOP lesions, including WWOP total area, total number, zonal distribution (Area and Number for R1 (posterior pole), R2 (mid-periphery), and R3 (far-periphery)), and global area ratio. Applied to 119 myopic eyes (64 with WWOP, 53.8%), the system revealed that over 96% of the lesion area resides outside the posterior pole, predominantly in the temporal and peripheral retina. Quantitative parameters showed significant positive correlations with axial length (e.g., total area: R = 0.41, P < 0.001; peripheral area R3: R = 0.40, P < 0.001; total number: R = 0.43, P < 0.001). This AI-powered system provides an objective, standardized, and high-throughput tool for WWOP evaluation but should currently be considered a research tool requiring further validation. Nevertheless, it holds strong potential to enhance clinical screening and risk stratification in myopia management.