Accurate brain tumour segmentation is a crucial step towards improving disease diagnosis and proper treatment planning. In this paper, we propose a deep-learning based method to segment a brain tumour into its subregions: whole tumour, tumour core and enhancing tumour. The proposed architecture is a 3D convolutional neural network based on a variant of the U-Net architecture of Ronneberger et al. [17] with three main modifications: (i) a heavy encoder, light decoder structure using residual blocks (ii) employment of dense blocks instead of skip connections, and (iii) utilization of self-ensembling in the decoder part of the network. The network was trained and tested using two different approaches: a multitask framework to segment all tumour subregions at the same time and a three-stage cascaded framework to segment one sub-region at a time. An ensemble of the results from both frameworks was also computed. To address the class imbalance issue, appropriate patch extraction was employed in a pre-processing step. The connected component analysis was utilized in the post-processing step to reduce false positive predictions. Experimental results on the BraTS20 validation dataset demonstrates that the proposed model achieved average Dice Scores of 0.90, 0.82, and 0.78 for whole tumour, tumour core and enhancing tumour respectively.

中文翻译：

---

使用级联 3D 密集连接的 U-net 进行脑肿瘤分割

准确的脑肿瘤分割是改善疾病诊断和正确治疗计划的关键一步。在本文中，我们提出了一种基于深度学习的方法，将脑肿瘤分割为其子区域：整个肿瘤、肿瘤核心和增强肿瘤。所提出的架构是基于 Ronneberger 等人的 U-Net 架构变体的 3D 卷积神经网络。[17] 有三个主要修改：（i）重编码器，使用残差块的轻解码器结构（ii）使用密集块而不是跳过连接，以及（iii）在网络的解码器部分使用自集成。该网络使用两种不同的方法进行训练和测试：同时分割所有肿瘤子区域的多任务框架和一次分割一个子区域的三级级联框架。还计算了来自两个框架的结果的集合。为了解决类别不平衡问题，在预处理步骤中采用了适当的补丁提取。在后处理步骤中利用连通分量分析来减少假阳性预测。BraTS20 验证数据集的实验结果表明，所提出的模型在整个肿瘤、肿瘤核心和增强肿瘤方面的平均 Dice 分数分别为 0.90、0.82 和 0.78。