DDU-Nets: Distributed Dense Model for 3D MRI Brain Tumor Segmentation
This work addresses brain tumor segmentation for medical imaging, presenting an incremental improvement in CNN efficiency.
The paper tackled brain tumor segmentation from 3D MRI by proposing DDU-Nets with distributed dense connections to enhance feature reuse, achieving effective results on the BraTS 2019 dataset with reduced computational cost.
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.