VesselView: A CNN for Segmentation of Vessels in High-Resolution Retinal Fundus Images
Retinal fundus imaging offers a noninvasive window into the eye’s microvasculature, critical for early detection of both ocular and systemic diseases. In this work, we introduce VesselView, a U-Net–inspired convolutional neural network designed for precise segmentation of retinal vessels in high-resolution fundus images. VesselView features double-convolution residual blocks with large kernels, a deepened bottleneck and skip connections. We conduct a fair comparative evaluation on the FIVES dataset at its full resolution (2048 × 2048), benchmarking VesselView against state-of-the-art models. The quantitative results demonstrate that VesselView achieves superior overall performance—as measured by the area under the ROC curve—with particularly strong results in glaucomatous and normal images. The qualitative comparative provides complementary evidence to the quantitative findings by showing that VesselView balances fewer missed vessels with more false positives than competing models, and validates the critical role of the chosen skip connections through an ablation study. These findings underscore the potential of specialized deep learning architectures for high-resolution retinal vessel segmentation.
Palabras clave:
Publicación: Libro
1779706777709
25 de mayo de 2026
/research/publications/vesselview-a-cnn-for-segmentation-of-vessels-in-high-resolution-retinal-fundus-images
Retinal fundus imaging offers a noninvasive window into the eye’s microvasculature, critical for early detection of both ocular and systemic diseases. In this work, we introduce VesselView, a U-Net–inspired convolutional neural network designed for precise segmentation of retinal vessels in high-resolution fundus images. VesselView features double-convolution residual blocks with large kernels, a deepened bottleneck and skip connections. We conduct a fair comparative evaluation on the FIVES dataset at its full resolution (2048 × 2048), benchmarking VesselView against state-of-the-art models. The quantitative results demonstrate that VesselView achieves superior overall performance—as measured by the area under the ROC curve—with particularly strong results in glaucomatous and normal images. The qualitative comparative provides complementary evidence to the quantitative findings by showing that VesselView balances fewer missed vessels with more false positives than competing models, and validates the critical role of the chosen skip connections through an ablation study. These findings underscore the potential of specialized deep learning architectures for high-resolution retinal vessel segmentation. - R. Santos-Mateos, A. Velev-Santos, Xosé M. Pardo - 10.1007/978-3-031-99568-2_21 - 9783031995675
publications_es