Fig. 1 | Scientific Data

Fig. 1

From: Leuven-Haifa High-Resolution Fundus Image Dataset for Retinal Blood Vessel Segmentation and Glaucoma Diagnosis

Fig. 1

Segmentation pipeline and suggestions for algorithm development. Raw fundus images are manually annotated in the Lirot.ai application by junior annotators and corrected by an expert annotator21. The resulting manual segmentations could serve as ground truth for vessel segmentation algorithm development and could be used for the computation of microvascular biomarkers, e.g. PVBM toolbox20. Together with clinical metadata these biomarkers carry the potential for disease screening and risk stratification, classification and monitoring algorithms.

Back to article page