EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automatic Detection of Microaneurysms in Fundus Images Using an Ensemble-Based Segmentation Method

Vidas Raudonis, Arturas Kairys iD, Rasa Verkauskiene, Jelizaveta Sokolovska iD, Goran Petrovski iD, Vilma Jurate Balciuniene, Vallo Volke

Sensors · 24.03.2023 · 23 (7) · 3431

DOI: 10.3390/s23073431 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 24.03.2023. Verfasst von Vidas Raudonis und 6 weiteren Autor:innen. Beteiligte Einrichtung: Automation Department, Kaunas University of Technology, 51368 Kaunas, Lithuania. Hauptthemen laut MEDLINE: Microaneurysm, Diabetic Retinopathy. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: EEA and Norway Grants, European Economic Area (EEA) Grants.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this study, a novel method for automatic microaneurysm detection in color fundus images is presented. The proposed method is based on three main steps: (1) image breakdown to smaller image patches, (2) inference to segmentation models, and (3) reconstruction of the predicted segmentation map from output patches. The proposed segmentation method is based on an ensemble of three individual deep networks, such as U-Net, ResNet34-UNet and UNet++. The performance evaluation is based on the calculation of the Dice score and IoU values. The ensemble-based model achieved higher Dice score (0.95) and IoU (0.91) values compared to other network architectures. The proposed ensemble-based model demonstrates the high practical application potential for detection of early-stage diabetic retinopathy in color fundus images.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MicroaneurysmDiabetic RetinopathyHumansFundus OculiImage Processing, Computer-Assisted

Stichwörter der Autor:innen

diabetic retinopathy (DR)encoder-decoder deep neural networkimage segmentationmicroaneurysms (MAs)

Autor:innen und Institutionen

  • Automation Department, Kaunas University of Technology, 51368 Kaunas, Lithuania.
  • Institute of Endocrinology, Lithuanian University of Health Sciences, 50140 Kaunas, Lithuania.
  • Faculty of Medicine, University of Latvia, 1004 Riga, Latvia.
  • Center of Eye Research and Innovative Diagnostics, Department of Ophthalmology, Oslo University Hospital and Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, 0372 Oslo, Norway.
  • Department of Ophthalmology, University of Split School of Medicine and University Hospital Centre, 21000 Split, Croatia.
  • Lithuanian University of Health Sciences, 44307 Kaunas, Lithuania.

ORCID-Profile: Arturas Kairys, Jelizaveta Sokolovska, Goran Petrovski

Förderung

  • EEA and Norway Grants
  • European Economic Area (EEA) Grants

Zitieren

APA

Raudonis, V., Kairys, A., Verkauskiene, R., Sokolovska, J., Petrovski, G., Balciuniene, V. J., & Volke, V. (2023). Automatic Detection of Microaneurysms in Fundus Images Using an Ensemble-Based Segmentation Method. Sensors, 23(7), 3431. https://doi.org/10.3390/s23073431

Vancouver

Raudonis V, Kairys A, Verkauskiene R, Sokolovska J, Petrovski G, Balciuniene VJ, et al. Automatic Detection of Microaneurysms in Fundus Images Using an Ensemble-Based Segmentation Method. Sensors. 2023;23(7):3431. doi:10.3390/s23073431

BibTeX
@article{Raudonis2023,
  author = {Raudonis, Vidas and Kairys, Arturas and Verkauskiene, Rasa and Sokolovska, Jelizaveta and Petrovski, Goran and Balciuniene, Vilma Jurate and Volke, Vallo},
  title = {Automatic Detection of Microaneurysms in Fundus Images Using an Ensemble-Based Segmentation Method},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {7},
  pages = {3431},
  doi = {10.3390/s23073431}
}
Kostenlos starten →