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

Despite the success of convolutional neural networks for 3D medical-imagesegmentation, the architectures currently used are still not robust enough tothe protocols of different scanners, and the variety of image properties theyproduce. Moreover, access to large-scale datasets with annotated regions ofinterest is scarce, and obtaining good results is thus difficult. To overcomethese challenges, we introduce IB-U-Nets, a novel architecture with inductivebias, inspired by the visual processing in vertebrates. With the 3D U-Net asthe base, we add two 3D residual components to the second encoder blocks. Theyprovide an inductive bias, helping U-Nets to segment anatomical structures from3D images with increased robustness and accuracy. We compared IB-U-Nets withstate-of-the-art 3D U-Nets on multiple modalities and organs, such as theprostate and spleen, using the same training and testing pipeline, includingdata processing, augmentation and cross-validation. Our results demonstrate thesuperior robustness and accuracy of IB-U-Nets, especially on small datasets, asis typically the case in medical-image analysis. IB-U-Nets source code andmodels are publicly available.

IB-U-Nets: Improving medical image segmentation tasks with 3D Inductive Biased kernels