Context Aware 3D UNet for Brain Tumor Segmentation

TL;DR

This paper introduces a modified 3D UNet architecture with densely connected blocks and residual-inception modules, enhancing feature reuse and multi-scale information extraction for improved brain tumor segmentation.

Contribution

The paper proposes a novel 3D UNet variant incorporating dense connections and residual-inception blocks to better utilize multi-contextual features for brain tumor segmentation.

Findings

Achieved DSC scores of 89.12% for whole tumor

Achieved DSC scores of 84.74% for tumor core

Achieved DSC scores of 79.12% for enhancing tumor

Abstract

Deep convolutional neural network (CNN) achieves remarkable performance for medical image analysis. UNet is the primary source in the performance of 3D CNN architectures for medical imaging tasks, including brain tumor segmentation. The skip connection in the UNet architecture concatenates features from both encoder and decoder paths to extract multi-contextual information from image data. The multi-scaled features play an essential role in brain tumor segmentation. However, the limited use of features can degrade the performance of the UNet approach for segmentation. In this paper, we propose a modified UNet architecture for brain tumor segmentation. In the proposed architecture, we used densely connected blocks in both encoder and decoder paths to extract multi-contextual information from the concept of feature reusability. In addition, residual-inception blocks (RIB) are used to extract the local and global information by merging features of different kernel sizes. We validate the proposed architecture on the multi-modal brain tumor segmentation challenge (BRATS) 2020 testing dataset. The dice (DSC) scores of the whole tumor (WT), tumor core (TC), and enhancing tumor (ET) are 89.12%, 84.74%, and 79.12%, respectively.