DDU-Nets: Distributed Dense Model for 3D MRI Brain Tumor Segmentation

• Main Authors: Zhang, Hanxiao, Li, Jingxiong, Shen, Mali, Wang, Yaqi, Yang, Guang-Zhong
• Format: Journal Article
• Language: English
• Published: 03-03-2020
• Subjects: Computer Science - Computer Vision and Pattern Recognition
• DOI: 10.48550/arxiv.2003.01337

Segmentation of brain tumors and their subregions remains a challenging task
due to their weak features and deformable shapes. In this paper, three patterns
(cross-skip, skip-1 and skip-2) of distributed dense connections (DDCs) are
proposed to enhance feature reuse and propagation of CNNs by constructing
tunnels between key layers of the network. For better detecting and segmenting
brain tumors from multi-modal 3D MR images, CNN-based models embedded with DDCs
(DDU-Nets) are trained efficiently from pixel to pixel with a limited number of
parameters. Postprocessing is then applied to refine the segmentation results
by reducing the false-positive samples. The proposed method is evaluated on the
BraTS 2019 dataset with results demonstrating the effectiveness of the DDU-Nets
while requiring less computational cost.