English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Fundus photography has routinely been used to document the presence andseverity of retinal degenerative diseases such as age-related maculardegeneration (AMD), glaucoma, and diabetic retinopathy (DR) in clinicalpractice, for which the fovea and optic disc (OD) are important retinallandmarks. However, the occurrence of lesions, drusen, and other retinalabnormalities during retinal degeneration severely complicates automaticlandmark detection and segmentation. Here we propose HBA-U-Net: a U-Netbackbone enriched with hierarchical bottleneck attention. The network consistsof a novel bottleneck attention block that combines and refines self-attention,channel attention, and relative-position attention to highlight retinalabnormalities that may be important for fovea and OD segmentation in thedegenerated retina. HBA-U-Net achieved state-of-the-art results on foveadetection across datasets and eye conditions (ADAM: Euclidean Distance (ED) of25.4 pixels, REFUGE: 32.5 pixels, IDRiD: 32.1 pixels), on OD segmentation forAMD (ADAM: Dice Coefficient (DC) of 0.947), and on OD detection for DR (IDRiD:ED of 20.5 pixels). Our results suggest that HBA-U-Net may be well suited forlandmark detection in the presence of a variety of retinal degenerativediseases.

U-Net with Hierarchical Bottleneck Attention for Landmark Detection in Fundus Images of the Degenerated Retina