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

With the thriving of deep learning, 3D Convolutional Neural Networks havebecome a popular choice in volumetric image analysis due to their impressive 3Dcontexts mining ability. However, the 3D convolutional kernels will introduce asignificant increase in the amount of trainable parameters. Considering thetraining data is often limited in biomedical tasks, a tradeoff has to be madebetween model size and its representational power. To address this concern, inthis paper, we propose a novel 3D Dense Separated Convolution (3D-DSC) moduleto replace the original 3D convolutional kernels. The 3D-DSC module isconstructed by a series of densely connected 1D filters. The decomposition of3D kernel into 1D filters reduces the risk of over-fitting by removing theredundancy of 3D kernels in a topologically constrained manner, while providingthe infrastructure for deepening the network. By further introducing nonlinearlayers and dense connections between 1D filters, the network's representationalpower can be significantly improved while maintaining a compact architecture.We demonstrate the superiority of 3D-DSC on volumetric image classification andsegmentation, which are two challenging tasks often encountered in biomedicalimage computing.

3D Dense Separated Convolution Module for Volumetric Image Analysis