VasFAZ-Net: A Hybrid Loss-Optimized E cientNet-scSE Network for Retinal Vasculature and FAZ Segmentation in OCTA

Simple item page
dc.contributor.authorRaja, Nisan Pranavah
dc.contributor.authorGopi, Varun P.
dc.contributor.authorFrancis, Bibin
dc.contributor.authorDevi, Chelli N.
dc.date.accessioned2026-11-02T05:57:35Z
dc.date.available2026-11-02T05:57:35Z
dc.date.issued2026-05-01
dc.description.abstractAccurate segmentation of ne retinal vessels and the Foveal Avascular Zone (FAZ) in OCTA images remains challenging due to low contrast, noise, and class imbalance. To address this, a lightweight deep learning framework called VasFAZ-Net is proposed, which uses an E cientNet-B0 encoder, a parallel spatial and channel squeeze-and-excitation (scSE) decoder, and a hybrid Dice-Focal loss function, trained on the OCTA-500 dataset with both 3 mm and 6 mm scans. The proposed model achieved Dice scores of up to 88.48% for vessel segmentation and 97.98% for FAZ segmentation with only 8.8 million parameters and an inference time of 18.2 ms per image. These results demonstrate that the proposed method provides accurate, computationally e cient, and clinically deployable multi-target OCTA segmentation for retinal disease analysis.
dc.identifier.issn15614042
dc.identifier.urihttps://doi.org/10.56415/csjm.v34.01
dc.identifier.urihttp://nitt.ndl.gov.in/handle/123456789/389
dc.language.isoen
dc.publisherInstitute of Mathematics and Computer Science
dc.relation.ispartofseriesomputer Science Journal of Moldova; Vol. 34/ No. 1
dc.titleVasFAZ-Net: A Hybrid Loss-Optimized E cientNet-scSE Network for Retinal Vasculature and FAZ Segmentation in OCTA
dc.typeArticle
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
VasFAZ-Net A Hybrid Loss-Optimized E cientNet-scSE Network for Retinal Vasculature and FAZ Segmentation in OCTA.pdf
Size:
746.79 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
licence.pdf
Size:
88.36 KB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Journal Articles