Morphology-aware distillation for lightweight retinal vessel segmentation across fundus photography and OCT angiography
Frontiers in Cell and Developmental Biology · 28.05.2026 · 14
DOI: 10.3389/fcell.2026.1825518 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Cell and Developmental Biology, online erschienen am 28.05.2026. Verfasst von Han Shu und 10 weiteren Autor:innen. Beteiligte Einrichtung: School of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, China. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Retinal vessel segmentation is critical for diagnosing ophthalmic and systemic diseases, yet deploying high-performance models in resource-constrained clinical settings remains a challenge. While Knowledge Distillation (KD) offers a solution for model compression, conventional KD methods often treat the U-Net as a generic feature extractor, neglecting the unique topological nature of vascular networks. This oversight frequently leads to “fractured” segmentation maps in student models, where fine capillaries and continuous vessel branches are lost.
Methods
In this paper, we advocate a task-specific, morphology-aware distillation strategy for lightweight retinal vessel segmentation. We introduce Morphology-Aware Reconstruction Distillation (MRD), a novel framework designed to transfer the teacher’s capability to reconstruct coherent vascular graphs rather than merely mimicking pixel statistics. Central to MRD is the Hierarchical Structure Fusion (HSF) module, a purpose-built unit that adaptively integrates multi-scale features using a tailored residual gating mechanism. By focusing on the decoder’s role as a topological reconstruction engine, HSF guides the student to learn how to synthesize continuous vessel structures from compressed representations. We validate our approach across two distinct modalities: Fundus Photography (FIVES and DRIVE datasets) and OCT Angiography (ROSE dataset).
Results
Extensive experiments demonstrate that our method significantly outperforms existing KD techniques. Notably, our distilled student model achieves diagnostic-level segmentation fidelity that surpasses massive teacher architectures like TransUNet, all while operating with a radically reduced parameter footprint.
Discussion
By robustly preserving delicate vascular morphology across diverse datasets, these results highlight the strong translational potential of our approach, providing a structurally reliable and highly efficient solution tailored for point-of-care clinical deployment. The code will be released.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, China.
- School of Information Science, Japan Advanced Institute of Science and Technology, Nomi, Japan.
- Department of Electronic Engineering, Tsinghua University, Beijing, China.
- Advanced Institute of So-Go-Chi (Convergence Knowledge) Informatics, Tohoku University, Sendai, Miyagi, Japan.
Zitieren
APA
Shu, H., Yuan, Q., Pan, Y., Kong, W., Feng, Y., Peng, C., Li, K., Nguyen, L. M., Racharak, T., Xin, J., & Sun, G. (2026). Morphology-aware distillation for lightweight retinal vessel segmentation across fundus photography and OCT angiography. Frontiers in Cell and Developmental Biology, 14. https://doi.org/10.3389/fcell.2026.1825518
Vancouver
Shu H, Yuan Q, Pan Y, Kong W, Feng Y, Peng C, et al. Morphology-aware distillation for lightweight retinal vessel segmentation across fundus photography and OCT angiography. Frontiers in Cell and Developmental Biology. 2026;14. doi:10.3389/fcell.2026.1825518
BibTeX
@article{Shu2026,
author = {Shu, Han and Yuan, Qi and Pan, Yizhi and Kong, Weikun and Feng, Yangyi and Peng, Cheng and Li, Kai and Nguyen, Le-Minh and Racharak, Teeradaj and Xin, Junyi and Sun, Guanqun},
title = {Morphology-aware distillation for lightweight retinal vessel segmentation across fundus photography and OCT angiography},
journal = {Frontiers in Cell and Developmental Biology},
year = {2026},
volume = {14},
doi = {10.3389/fcell.2026.1825518}
}
EUVIMED