3D-DDA: 3D Dual-Domain Attention for Brain Tumor Segmentation

Accurate brain tumor segmentation plays an essential role in the diagnosis process. However, there are challenges due to the variety of tumors in low contrast, morphology, location, annotation bias, and imbalance among tumor regions. This work proposes a novel 3D dual-domain attention module to lear...

Full description

Saved in:
Bibliographic Details
Published in2023 IEEE International Conference on Image Processing (ICIP) pp. 3215 - 3219
Main Authors Do, Nhu-Tai, Vo-Thanh, Hoang-Son, Nguyen-Quynh, Tram-Tran, Kim, Soo-Hyung
Format Conference Proceeding
LanguageEnglish
Published IEEE 08.10.2023
Subjects
3D dual-domain attention
Annotations
Brain tumor segmentation
Image coding
Image segmentation
medical image analysis
Morphology
Three-dimensional displays
Tumors
Uncertainty
Online AccessGet full text

Cover

More Information
Summary:Accurate brain tumor segmentation plays an essential role in the diagnosis process. However, there are challenges due to the variety of tumors in low contrast, morphology, location, annotation bias, and imbalance among tumor regions. This work proposes a novel 3D dual-domain attention module to learn local and global information in spatial and context domains from encoding feature maps in Unet. Our attention module generates refined feature maps from the enlarged reception field at every stage by attention mechanisms and residual learning to focus on complex tumor regions. Our experiments on BraTS 2018 have demonstrated superior performance compared to existing state-of-the-art methods.
DOI:10.1109/ICIP49359.2023.10222602