Topology-Aware Segmentation Using Discrete Morse Theory
Topology-Aware Segmentation Using Discrete Morse Theory
复制标题
DOI：
--
复制
发表时间：
2021-03
期刊：
影响因子：
--
通讯作者：
Xiaoling Hu;Yusu Wang;Fuxin Li;D. Samaras;Chao Chen
中科院分区：
文献类型：
--
作者：
Xiaoling Hu;Yusu Wang;Fuxin Li;D. Samaras;Chao Chen
在从自然和生物医学图像中分割精细尺度结构时，逐像素精度并不是唯一需要关注的指标。拓扑结构的正确性，如血管连通性和膜闭合性，对于下游分析任务至关重要。在本文中，我们提出了一种新的方法来训练深度图像分割网络以获得更好的拓扑精度。特别是，利用离散莫尔斯理论（DMT）的力量，我们确定了全局结构，包括1D骨架和2D补丁，这对拓扑精度很重要。使用基于这些全局结构的新损失进行训练，网络性能显着提高，特别是在拓扑结构具有挑战性的位置附近（如连接和膜的薄弱点）。在不同的数据集上，我们的方法在DICE得分和拓扑指标上都取得了优异的性能。
In the segmentation of fine-scale structures from natural and biomedical images, per-pixel accuracy is not the only metric of concern. Topological correctness, such as vessel connectivity and membrane closure, is crucial for downstream analysis tasks. In this paper, we propose a new approach to train deep image segmentation networks for better topological accuracy. In particular, leveraging the power of discrete Morse theory (DMT), we identify global structures, including 1D skeletons and 2D patches, which are important for topological accuracy. Trained with a novel loss based on these global structures, the network performance is significantly improved especially near topologically challenging locations (such as weak spots of connections and membranes). On diverse datasets, our method achieves superior performance on both the DICE score and topological metrics.