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/

Zendy Plus

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

Zendy Tools

Zendy Open

One of the key drawbacks of 3D convolutional neural networks for segmentationis their memory footprint, which necessitates compromises in the networkarchitecture in order to fit into a given memory budget. Motivated by theRevNet for image classification, we propose a partially reversible U-Netarchitecture that reduces memory consumption substantially. The reversiblearchitecture allows us to exactly recover each layer's outputs from thesubsequent layer's ones, eliminating the need to store activations forbackpropagation. This alleviates the biggest memory bottleneck and enables verydeep (theoretically infinitely deep) 3D architectures. On the BraTS challengedataset, we demonstrate substantial memory savings. We further show that thefreed memory can be used for processing the whole field-of-view (FOV) insteadof patches. Increasing network depth led to higher segmentation accuracy whilegrowing the memory footprint only by a very small fraction, thanks to thepartially reversible architecture.

A Partially Reversible U-Net for Memory-Efficient Volumetric Image Segmentation