EUVIMEDEUVIMED entdecken
Open Access

Robust semi-automatic vessel tracing in the human retinal image by an instance segmentation neural network

Siyi Chen iD, Linh Hoang iD, Amir H Kashani iD, Ji Yi iD

Science Advances · 2025

DOI: 10.1126/sciadv.ado8268 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, 2025. Verfasst von Siyi Chen und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21231, USA. Hauptthemen laut MEDLINE: Neural Networks, Computer, Retinal Vessels, Image Processing, Computer-Assisted, Retina. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NEI NIH HHS, NINDS NIH HHS, NIA NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Vasculature morphology and hierarchy are essential for blood perfusion. Human retinal circulation is an intricate vascular system emerging and remerging at the optic nerve head (ONH). Tracing retinal vascular branching from ONH can allow detailed morphological quantification, and yet remains a challenging task. We presented a robust semi-automatic vessel tracing algorithm on human fundus images by an instance segmentation neural network (InSegNN). InSegNN separates and labels individual vascular trees and enables tracing each tree throughout its branching. We have three strategies to improve robustness and accuracy: pseudotemporal learning, spatial multisampling, and dynamic probability map. We achieved 83% specificity, 50% improvement in symmetric best dice (SBD) compared to literature, and outperformed baseline U-net, and achieved 91% precision with 71% sensitivity. We have demonstrated tracing individual vessel trees from fundus images, and simultaneously retain vessel hierarchy information. InSegNN paves a way for subsequent analysis of vascular morphology in relation to retinal diseases.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerRetinal VesselsImage Processing, Computer-AssistedRetinaHumansAlgorithmsOptic DiskFundus Oculi

Autor:innen und Institutionen

  • Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21231, USA.
  • Department of Ophthalmology, Johns Hopkins University, School of Medicine, Baltimore, MD 21231, USA.

ORCID-Profile: Siyi Chen, Linh Hoang, Amir H Kashani, Ji Yi

Förderung

  • NEI NIH HHS
  • NINDS NIH HHS
  • NIA NIH HHS

Zitieren

APA

Chen, S., Hoang, L., Kashani, A. H., & Yi, J. (2025). Robust semi-automatic vessel tracing in the human retinal image by an instance segmentation neural network. Science Advances. https://doi.org/10.1126/sciadv.ado8268

Vancouver

Chen S, Hoang L, Kashani AH, Yi J. Robust semi-automatic vessel tracing in the human retinal image by an instance segmentation neural network. Science Advances. 2025. doi:10.1126/sciadv.ado8268

BibTeX
@article{Chen2025,
  author = {Chen, Siyi and Hoang, Linh and Kashani, Amir H and Yi, Ji},
  title = {Robust semi-automatic vessel tracing in the human retinal image by an instance segmentation neural network},
  journal = {Science Advances},
  year = {2025},
  doi = {10.1126/sciadv.ado8268}
}
Kostenlos starten →