AI-Based Fundus Screening of Teachers: Fundus Tessellated Density as a Myopia Biomarker and Occupational Associations

Purpose This study was to investigate the practical application of artificial intelligence (AI)-based fundus screening system for ophthalmic health management in the teacher population, specifically aiming to explore the etiology and potential pathogenesis of high myopia-associated fundus lesion progression. Methods A randomized of primary and secondary school teachers from Beijing, provided color fundus images and health examination data. The study innovatively employed AI for assisted diagnosis and analysis. Results The mean spherical equivalent (SE) was − 2.43 ± 2.94 D. Univariate analysis identified significant correlations between SE and factors such as age, teaching subject, and near work duration. Quantitative analysis revealed that fundus tessellated density (FTD) was highest in the nasal macular quadrant. Both overall and nasal FTD within the 3 mm macular region effectively discriminated high myopia (SE ≤ -6.00 D), with area under the curve values of 0.7788 and 0.7631, respectively. Furthermore, teachers of literature subjects exhibited significantly lower SE and higher FTD, while prolonged near work (≥ 1 hour) at 30 cm was associated with increased FTD. Finally, FTD positively correlated with systemic immune-inflammatory indices (neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio), which were elevated in high myopia. Conclusion This study demonstrates the characteristics of the progression of myopia and fundus lesions in the teacher population through AI-based fundus screening system. AI-quantified FTD is a valuable biomarker for high myopia, is influenced by occupational near work, and is associated with a systemic pro-inflammatory state.