TotalSegmentator: robust segmentation of 104 anatomical structures in CT images
TotalSegmentator: robust segmentation of 104 anatomical structures in CT images
复制标题
DOI：
10.48550/arxiv.2208.05868
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Jakob Wasserthal;M. Meyer;H. Breit;Joshy Cyriac;Shan Yang;Martin Segeroth
中科院分区：
文献类型：
--
作者：
Jakob Wasserthal;M. Meyer;H. Breit;Joshy Cyriac;Shan Yang;Martin Segeroth
在这项工作中，我们重点研究了（全身）CT图像中多个解剖结构的自动分割。针对这一任务存在许多分割算法。然而，在大多数情况下，他们会遇到3个问题：它们难以使用（代码和数据不是公开可用的或难以使用）。2. 他们没有泛化（通常训练数据集只包含非常干净的图像，而不反映临床常规中发现的图像分布），2。该算法只能分割一个解剖结构。对于更多的结构，必须使用几种算法，这增加了建立系统所需的工作量。在这项工作中，我们发布了一个新的数据集和分割工具包，解决了所有这三个问题：在1204个CT图像中，我们分割了104个解剖结构（27个器官，59个骨骼，10个肌肉，8个血管），涵盖了大多数用例的大多数相关类别。我们展示了一个改进的工作流程，用于创建地面真实分割，它将该过程加快了10倍以上。CT图像是从临床常规中随机抽取的，因此代表了一个真实世界的数据集，可以推广到临床应用。该数据集包含广泛的不同病理、扫描仪、序列和位点。最后，我们在这个新的数据集上训练了一个分割算法。我们称之为
In this work we focus on automatic segmentation of multiple anatomical structures in (whole body) CT images. Many segmentation algorithms exist for this task. However, in most cases they suffer from 3 problems: 1. They are difficult to use (the code and data is not publicly available or difficult to use). 2. They do not generalize (often the training dataset was curated to only contain very clean images which do not reflect the image distribution found during clinical routine), 3. The algorithm can only segment one anatomical structure. For more structures several algorithms have to be used which increases the effort required to set up the system. In this work we publish a new dataset and segmentation toolkit which solves all three of these problems: In 1204 CT images we segmented 104 anatomical structures (27 organs, 59 bones, 10 muscles, 8 vessels) covering a majority of relevant classes for most use cases. We show an improved workflow for the creation of ground truth segmentations which speeds up the process by over 10x. The CT images were randomly sampled from clinical routine, thus representing a real world dataset which generalizes to clinical application. The dataset contains a wide range of different pathologies, scanners, sequences and sites. Finally, we train a segmentation algorithm on this new dataset. We call this