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

Liver cancer is one of the leading causes of cancer death. To assist doctorsin hepatocellular carcinoma diagnosis and treatment planning, an accurate andautomatic liver and tumor segmentation method is highly demanded in theclinical practice. Recently, fully convolutional neural networks (FCNs),including 2D and 3D FCNs, serve as the back-bone in many volumetric imagesegmentation. However, 2D convolutions can not fully leverage the spatialinformation along the third dimension while 3D convolutions suffer from highcomputational cost and GPU memory consumption. To address these issues, wepropose a novel hybrid densely connected UNet (H-DenseUNet), which consists ofa 2D DenseUNet for efficiently extracting intra-slice features and a 3Dcounterpart for hierarchically aggregating volumetric contexts under the spiritof the auto-context algorithm for liver and tumor segmentation. We formulatethe learning process of H-DenseUNet in an end-to-end manner, where theintra-slice representations and inter-slice features can be jointly optimizedthrough a hybrid feature fusion (HFF) layer. We extensively evaluated ourmethod on the dataset of MICCAI 2017 Liver Tumor Segmentation (LiTS) Challengeand 3DIRCADb Dataset. Our method outperformed other state-of-the-arts on thesegmentation results of tumors and achieved very competitive performance forliver segmentation even with a single model.

H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes