Key points are not available for this paper at this time.
Background: Fundus tessellation (FT) is increasingly recognized as an early structural manifestation of retinal and choroidal remodeling in myopia, reflecting initial changes associated with axial elongation. With advances in artificial intelligence (AI), particularly deep learning-based image analysis, quantitative assessment of FT has emerged as a promising approach for objective and scalable evaluation in myopia research. Methods: This review provides an integrative overview of recent studies applying AI-assisted and quantitative image-analysis approaches to fundus tessellation assessment. Relevant literature was identified through a structured search of major biomedical databases, focusing on methodologies for FT and fundus tessellation density (FTD) quantification and their reported associations with clinical ocular parameters in myopia. Results: Across multiple cohorts, AI-derived FTD consistently showed associations with axial length elongation, choroidal thinning, and increasing myopia severity. Common analytical frameworks involved region-of-interest definition, image normalization, and supervised deep learning-based segmentation. Several studies further reported spatial heterogeneity in FTD distribution and its relationship with peripapillary and macular alterations, supporting the potential role of FT-related metrics as quantitative imaging biomarkers for myopic structural change. Conclusion: AI-driven quantification of fundus tessellation represents a methodological advancement in myopia research by enhancing objectivity and scalability in retinal image analysis. These approaches may facilitate early risk stratification and hold promise for future longitudinal assessment of myopia.
Guo et al. (2026) studied this question.