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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Optical coherence tomography (OCT) is a noninvasive imaging modality whichcan be used to obtain depth images of the retina. The changing layerthicknesses can thus be quantified by analyzing these OCT images, moreoverthese changes have been shown to correlate with disease progression in multiplesclerosis. Recent automated retinal layer segmentation tools use machinelearning methods to perform pixel-wise labeling and graph methods to guaranteethe layer hierarchy or topology. However, graph parameters like distance andsmoothness constraints must be experimentally assigned by retinal region andpathology, thus degrading the flexibility and time efficiency of the wholeframework. In this paper, we develop cascaded deep networks to provide atopologically correct segmentation of the retinal layers in a single feedforward propagation. The first network (S-Net) performs pixel-wise labeling andthe second regression network (R-Net) takes the topologically unconstrainedS-Net results and outputs layer thicknesses for each layer and each position.Relu activation is used as the final operation of the R-Net which guaranteesnon-negativity of the output layer thickness. Since the segmentation boundaryposition is acquired by summing up the corresponding non-negative layerthicknesses, the layer ordering (i.e., topology) of the reconstructedboundaries is guaranteed even at the fovea where the distances betweenboundaries can be zero. The R-Net is trained using simulated masks and thus canbe generalized to provide topology guaranteed segmentation for other layeredstructures. This deep network has achieved comparable mean absolute boundaryerror (2.82 {\mu}m) to state-of-the-art graph methods (2.83 {\mu}m).

Topology guaranteed segmentation of the human retina from OCT using convolutional neural networks