3D OCTA Microvascular Annotations

Two OCTA datasets were employed in this study. The first is a clinical dataset that includes scans from both healthy individuals and patients with diabetic retinopathy, acquired at a typical macular field of view (3 × 3 mm). The second dataset consists of annotated scans from healthy subjects with the same imaging resolution. These datasets were used to validate the proposed method in terms of segmentation accuracy, robustness, and generalizability across different imaging conditions.