Geometry Before Volume: A Review of Morphology-Aware Learning for Pulmonary Vessel Segmentation

Pulmonary vessel segmentation is often evaluated as a binary labeling problem, yet the anatomy to be recovered is a sparse, branching, multiscale network. A small break in a distal artery can have little influence on regional overlap while disrupting an entire downstream subtree; a slight boundary bias in a large vessel can dominate the voxel count while leaving clinically important connectivity unchanged. This mismatch has encouraged a shift from architecture-only innovation toward morphology-aware learning, in which centerlines, boundary distance, thickness, and continuity become explicit training signals. This review traces that shift from classical vessel enhancement and encoder-decoder segmentation to topology-aware losses, geometry-conditioned convolutions, and adapted foundation models. MorVess is examined as a recent synthesis of these ideas because it jointly predicts vessel masks, distance maps, and thickness maps while introducing limited three-dimensional context into a two-dimensional Segment Anything Model encoder. The analysis identifies three recurring design tensions: local detail versus global tree structure, full volumetric context versus computational economy, and general-purpose representation versus anatomy-specific supervision. Evidence suggests that geometric priors can improve small-branch recovery and diameter consistency, but their value depends on annotation quality, voxel spacing, skeleton stability, and evaluation beyond aggregate Dice. The review concludes with design principles for morphology-aware pulmonary vessel segmentation and with research needs for external validation, uncertainty-aware topology, and clinically grounded morphometry.

1. Mao, F., Chen, Y., Wu, B., Lin, L., Dai, J., Li, Z., ... & Qin, F. (2026). MorVess: Morphology-Aware Pulmonary Vessel Segmentation Network. arXiv preprint arXiv:2606.24214.
2. Chu, Y., Luo, G., Zhou, L., Cao, S., Ma, G., Meng, X., ... & Gao, X. (2025). Deep learning-driven pulmonary artery and vein segmentation reveals demography-associated vasculature anatomical differences. Nature Communications, 16, 2262.
3. Cicek, O., Abdulkadir, A., Lienkamp, S. S., Brox, T., & Ronneberger, O. (2016). 3D U-Net: Learning dense volumetric segmentation from sparse annotation. In Medical Image Computing and Computer-Assisted Intervention - MICCAI 2016 (pp. 424-432). Springer.
4. Isensee, F., Jaeger, P. F., Kohl, S. A. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation. Nature Methods, 18(2), 203-211.
5. Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., ... & Girshick, R. (2023). Segment Anything. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 4015-4026).
6. Luo, G., Wang, K., Liu, J., Li, S., Liang, X., Li, X., ... & Gao, X. (2023). Efficient automatic segmentation for multi-level pulmonary arteries: The PARSE challenge. arXiv preprint arXiv:2304.03708.
7. Ma, J., He, Y., Li, F., Han, L., You, C., Wang, B., ... & Wang, B. (2024). Segment anything in medical images. Nature Communications, 15, 654.
8. Moccia, S., De Momi, E., El Hadji, S., & Mattos, L. S. (2018). Blood vessel segmentation algorithms - review of methods, datasets and evaluation metrics. Computer Methods and Programs in Biomedicine, 158, 71-91.
9. Qi, Y., He, Y., Qi, X., Zhang, Y., & Yang, G. (2023). Dynamic Snake Convolution Based on Topological Geometric Constraints for Tubular Structure Segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 6070-6079).
10. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015 (pp. 234-241). Springer.
11. Shit, S., Paetzold, J. C., Sekuboyina, A., Ezhov, I., Unger, A., Zhylka, A., ... & Menze, B. H. (2021). clDice - A novel topology-preserving loss function for tubular structure segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 16560-16569).
12. Wang, Y., Wei, X., Liu, F., Chen, J., Zhou, Y., Shen, W., Fishman, E. K., & Yuille, A. L. (2020). Deep Distance Transform for tubular structure segmentation in CT scans. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 3833-3842).
13. Wu, Y., Qi, S., Wang, M., Zhao, S., Pang, H., Xu, J., Bai, L., & Ren, H. (2023). Transformer-based 3D U-Net for pulmonary vessel segmentation and artery-vein separation from CT images. Medical & Biological Engineering & Computing, 61(10), 2649-2663.
14. Xia, L., Zhang, H., Wu, Y., Song, R., Ma, Y., Mou, L., ... & Zhao, Y. (2022). 3D vessel-like structure segmentation in medical images by an edge-reinforced network. Medical Image Analysis, 82, 102581.