VoteNet: A Deep Learning Label Fusion Method for Multi-Atlas Segmentation.
VoteNet: A Deep Learning Label Fusion Method for Multi-Atlas Segmentation.
复制标题
DOI：
10.1007/978-3-030-32248-9_23
复制
发表时间：
2019-10
期刊：
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
深度学习 (DL) 方法对于许多医学图像分割任务来说都是最先进的。它们具有许多优点：可以针对特定任务进行训练，测试时计算速度很快，并且分割质量通常很高。相比之下，以前流行的多图集分割（MAS）方法相对较慢（因为它们依赖于昂贵的配准），并且尽管已经提出了复杂的标签融合策略，DL方法通常优于MAS。在这项工作中，我们提出了一种基于深度学习的标签融合策略（VoteNet），该策略在本地选择一组可靠的图集，然后通过多数投票将其标签融合。 3D 脑 MRI 数据实验表明，通过选择良好的初始图集集，使用 VoteNet 的 MAS 显着优于许多其他标签融合策略以及直接 DL 分割方法。我们还提供了通过我们的方法可实现的性能上限的实验分析。虽然在实践中不太可能实现，但这一界限表明进一步提高性能的空间。最后，为了解决标准 MAS 的运行时缺点，我们所有的结果都使用快速 DL 注册方法。
Deep learning (DL) approaches are state-of-the-art for many medical image segmentation tasks. They offer a number of advantages: they can be trained for specific tasks, computations are fast at test time, and segmentation quality is typically high. In contrast, previously popular multi-atlas segmentation (MAS) methods are relatively slow (as they rely on costly registrations) and even though sophisticated label fusion strategies have been proposed, DL approaches generally outperform MAS. In this work, we propose a DL-based label fusion strategy (VoteNet) which locally selects a set of reliable atlases whose labels are then fused via plurality voting. Experiments on 3D brain MRI data show that by selecting a good initial atlas set MAS with VoteNet significantly outperforms a number of other label fusion strategies as well as a direct DL segmentation approach. We also provide an experimental analysis of the upper performance bound achievable by our method. While unlikely achievable in practice, this bound suggests room for further performance improvements. Lastly, to address the runtime disadvantage of standard MAS, all our results make use of a fast DL registration approach.