EUVIMEDEUVIMED entdecken
Crossref

3DChoroidSwin: advancing 3D choroid segmentation in OCT images through Swin Transformer and morphological guidance

Takayuki Okamoto, Shingo Tamachi, Takehito Iwase, Tomohiro Niizawa, Yuto Kawamata, Hirotaka Yokouchi, Takayuki Baba, Hideaki Haneishi

Optics Express · 2025

DOI: 10.1364/oe.541344 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Optics Express, 2025. Verfasst von Takayuki Okamoto und 7 weiteren Autor:innen. Hauptthemen laut MEDLINE: Tomography, Optical Coherence, Choroid, Imaging, Three-Dimensional, Neural Networks, Computer. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: Japan Society for the Promotion of Science.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The choroid is a dense vascular layer that lies between the retina and the sclera and contributes to the blood supply of the outer retina. In recent years, optical coherence tomography (OCT), which enables non-destructive acquisition of cross-sectional images of the choroid, has revealed the relationship between morphological changes in the choroid and eye diseases. In this context, automatic and accurate segmentation of OCT images is critical, but many existing methods face challenges, as they 1) rely on convolutional neural network (CNN)-based architectures, which struggle to capture long-range dependencies, and 2) primarily focus on two-dimensional OCT images and thus have difficulty identifying the complex three-dimensional (3D) structure of the choroid. In this study, we propose an automatic choroid segmentation method, 3DChoroidSwin, which incorporates 3D CNN and 3D Swin Transformer frameworks, achieving both short- and long-distance learning. Furthermore, our method uses a combined loss function that includes the boundary loss, which leverages morphological information, achieving shape-aware training and decreasing unnatural false positives. Experimental results using clinical data demonstrate that the proposed method outperforms comparison methods, delivering performance comparable to ground truth; moreover, it achieves smooth and continuous 3D segmentation with reduced segmentation errors at the choroid margins.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Tomography, Optical CoherenceChoroidImaging, Three-DimensionalNeural Networks, ComputerHumansAlgorithms

Förderung

  • Japan Society for the Promotion of Science

Zitieren

APA

Okamoto, T., Tamachi, S., Iwase, T., Niizawa, T., Kawamata, Y., Yokouchi, H., Baba, T., & Haneishi, H. (2025). 3DChoroidSwin: advancing 3D choroid segmentation in OCT images through Swin Transformer and morphological guidance. Optics Express. https://doi.org/10.1364/oe.541344

Vancouver

Okamoto T, Tamachi S, Iwase T, Niizawa T, Kawamata Y, Yokouchi H, et al. 3DChoroidSwin: advancing 3D choroid segmentation in OCT images through Swin Transformer and morphological guidance. Optics Express. 2025. doi:10.1364/oe.541344

BibTeX
@article{Okamoto2025,
  author = {Okamoto, Takayuki and Tamachi, Shingo and Iwase, Takehito and Niizawa, Tomohiro and Kawamata, Yuto and Yokouchi, Hirotaka and Baba, Takayuki and Haneishi, Hideaki},
  title = {3DChoroidSwin: advancing 3D choroid segmentation in OCT images through Swin Transformer and morphological guidance},
  journal = {Optics Express},
  year = {2025},
  doi = {10.1364/oe.541344}
}
Kostenlos starten →