Data-Efficient Transformer Model for Retinal Artery-Vein Segmentation in Landmark-Deficient OCT Angiography

Accurately differentiating arteries from veins in highly magnified foveal 3 × 3-mm² macular OCT angiography (OCTA) is critical for quantifying retinal vascular disease. However, existing models require macrovascular landmarks absent at this field of view, hindering vessel-specific analysis in clinical imaging. We developed and validated a data-efficient, 2-stage transformer for artery-vein (AV) segmentation in landmark-deficient macular OCTA and evaluated whether the resulting vessel maps could identify retinal vein occlusion (RVO) biomarkers without disease-specific retraining.

Cross-sectional model development and validation study.

The development cohort comprised 17 eyes (29 scans) from 9 healthy adult volunteers (mean age, 32 years). A subsequent RVO pilot cohort included 9 foveal OCTA scans from patients with confirmed RVO, compared against 15 healthy control scans.

Independent 2-stage Swin-Unet transformer trained on sparse annotations from 12 eyes, applied without retraining to 5 test eyes and the RVO pilot cohort.

Artery-vein segmentation Dice coefficients, area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. Clinical RVO end points included the major vein-to-artery sinuosity ratio and the 10th-percentile artery radius.

The framework demonstrated exceptional internal validation and crucially, when applied to the external RVO cohort, the model successfully identified established pathophysiological hallmarks. Stage 1 global segmentation achieved AUC 0.973 and Dice 88.6%. Stage 2 AV classification achieved Dice 96.3% (arteries) and 96.1% (veins). End-to-end system Dice coefficients were 87.3% and 84.1% for arteries and veins, respectively. Applied to RVO without retraining, the model identified significantly elevated major vein/artery sinuosity (1.041 vs. 1.010; P = 0.007; AUC, 0.837) and reduced 10th-percentile artery radius (13.2 vs. 15.4 μm; P = 0.002; AUC, 0.885).

This data-efficient framework enables high-fidelity AV segmentation in landmark-deficient OCTA. By combining conventional methods with newer transformer models and decoupling structural segmentation from classification, the model identified biologically consistent biomarkers in RVO from small sets of scans. This supports highly scalable automated vessel specific quantification in clinical macular imaging; these pilot findings are hypothesis-generating and warrant prospective validation.

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.