Topology-Preserving Deep Image Segmentation
Topology-Preserving Deep Image Segmentation
复制标题
DOI：
--
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Xiaoling Hu;Fuxin Li;D. Samaras;Chao Chen
中科院分区：
文献类型：
--
作者：
Xiaoling Hu;Fuxin Li;D. Samaras;Chao Chen
分割算法容易在细尺度结构上产生拓扑错误，例如，断开连接。我们提出了一种新的方法，学习与正确的拓扑结构分割。特别地，我们设计了一个连续值损失函数，该函数强制分割具有与地面实况相同的拓扑，即，具有相同的贝蒂数提出的拓扑保持损失函数是可微的，我们将其纳入深度神经网络的端到端训练中。我们的方法取得了更好的性能的贝蒂数错误，这直接占拓扑正确性。它还在其他拓扑相关度量上表现优异，例如，调整后的兰德指数与信息的变异性。我们说明了所提出的方法在广泛的自然和生物医学数据集上的有效性。
Segmentation algorithms are prone to make topological errors on fine-scale structures, e.g., broken connections. We propose a novel method that learns to segment with correct topology. In particular, we design a continuous-valued loss function that enforces a segmentation to have the same topology as the ground truth, i.e., having the same Betti number. The proposed topology-preserving loss function is differentiable and we incorporate it into end-to-end training of a deep neural network. Our method achieves much better performance on the Betti number error, which directly accounts for the topological correctness. It also performs superiorly on other topology-relevant metrics, e.g., the Adjusted Rand Index and the Variation of Information. We illustrate the effectiveness of the proposed method on a broad spectrum of natural and biomedical datasets.