Pixel-Level Segmentation of Retinal Breaks in Ultra-Widefield Fundus Images with a PraNet-Based Machine Learning Model
Sensors · 2025
DOI: 10.3390/s25185862 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Takuya Takayama und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ophthalmology, Jichi Medical University, Shimotsuke, Tochigi 329-0498, Japan. Hauptthemen laut MEDLINE: Machine Learning, Fundus Oculi, Retina, Image Processing, Computer-Assisted. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Society for the Promotion of Science, JSPS KAKENHI.
Abstract (Originalsprache)
Retinal breaks are critical lesions that can cause retinal detachment and vision loss if not detected and treated early. Automated, accurate delineation of retinal breaks in ultra-widefield fundus (UWF) images remains challenging. In this study, we developed and validated a deep learning segmentation model based on the PraNet architecture to localize retinal breaks in break-positive cases. We trained and evaluated the model using a dataset comprising 34,867 UWF images of 8083 cases. Performance was assessed using image-level segmentation metrics, including accuracy, precision, recall, Intersection over Union (IoU), dice score, and centroid distance score. The model achieved an accuracy of 0.996, precision of 0.635, recall of 0.756, IoU of 0.539, dice score of 0.652, and centroid distance score of 0.081. To our knowledge, this is the first study to present pixel-level segmentation of retinal breaks in UWF images using deep learning. The proposed PraNet-based model showed high accuracy and robust segmentation performance, highlighting its potential for clinical application.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Ophthalmology, Jichi Medical University, Shimotsuke, Tochigi 329-0498, Japan.
- DeepEyeVision Inc., Shimotsuke, Tochigi 329-0498, Japan.
- Department of Ophthalmology, Yokohama City University, Kanagawa 232-0023, Japan.
- Retina Research Group, Singapore Eye Research Institute, Singapore 168751, Singapore.
- Center for Cyber Medicine Research, Tsukuba University, Ibaraki 305-8577, Japan.
ORCID-Profile: Takuya Takayama, Toshikatsu Kaburaki, Hidenori Takahashi
Förderung
- Japan Society for the Promotion of Science
- JSPS KAKENHI
Zitieren
APA
Takayama, T., Uto, T., Tsuge, T., Kondo, Y., Tampo, H., Chiba, M., Kaburaki, T., Yanagi, Y., & Takahashi, H. (2025). Pixel-Level Segmentation of Retinal Breaks in Ultra-Widefield Fundus Images with a PraNet-Based Machine Learning Model. Sensors. https://doi.org/10.3390/s25185862
Vancouver
Takayama T, Uto T, Tsuge T, Kondo Y, Tampo H, Chiba M, et al. Pixel-Level Segmentation of Retinal Breaks in Ultra-Widefield Fundus Images with a PraNet-Based Machine Learning Model. Sensors. 2025. doi:10.3390/s25185862
BibTeX
@article{Takayama2025,
author = {Takayama, Takuya and Uto, Tsubasa and Tsuge, Taiki and Kondo, Yusuke and Tampo, Hironobu and Chiba, Mayumi and Kaburaki, Toshikatsu and Yanagi, Yasuo and Takahashi, Hidenori},
title = {Pixel-Level Segmentation of Retinal Breaks in Ultra-Widefield Fundus Images with a PraNet-Based Machine Learning Model},
journal = {Sensors},
year = {2025},
doi = {10.3390/s25185862}
}
EUVIMED