AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem?
AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem?
复制标题
DOI：
10.1109/tpami.2021.3100536
复制
发表时间：
2022-10-01
影响因子：
23.6
通讯作者：
Yang, Xiaoping
中科院分区：
文献类型：
--
作者：
Ma, Jun;Zhang, Yao;Yang, Xiaoping
随着深度学习空前的发展，主要腹部器官的自动分割似乎已成为一个已解决的问题，因为最先进的（SOTA）方法已经在许多基准数据集上取得了与评估者间变异性相当的结果。然而，大多数现有的腹部数据集仅包含单中心、单阶段、单供应商或单疾病病例，并且尚不清楚优异的性能是否可以推广到不同的数据集。本文提出了一个大型且多样化的腹部 CT 器官分割数据集，称为 AbdomenCT-1K，包含来自 12 个医疗中心的 1000 多个 (1K) CT 扫描，包括多阶段、多供应商和多疾病病例。此外，我们对肝脏、肾脏、脾脏和胰腺分割进行了大规模研究，揭示了 SOTA 方法尚未解决的分割问题，例如对不同医疗中心、阶段和未见疾病的泛化能力有限。为了推进未解决的问题，我们进一步建立了全监督、半监督、弱监督和持续学习的四个器官分割基准，这是目前具有挑战性和活跃的研究课题。因此，我们为每个基准开发了一种简单有效的方法，可以用作开箱即用的方法和强大的基线。我们相信 AbdomenCT-1K 数据集将促进未来对临床适用的腹部器官分割方法的深入研究。
With the unprecedented developments in deep learning, automatic segmentation of main abdominal organs seems to be a solved problem as state-of-the-art (SOTA) methods have achieved comparable results with inter-rater variability on many benchmark datasets. However, most of the existing abdominal datasets only contain single-center, single-phase, single-vendor, or single-disease cases, and it is unclear whether the excellent performance can generalize on diverse datasets. This paper presents a large and diverse abdominal CT organ segmentation dataset, termed AbdomenCT-1K, with more than 1000 (1K) CT scans from 12 medical centers, including multi-phase, multi-vendor, and multi-disease cases. Furthermore, we conduct a large-scale study for liver, kidney, spleen, and pancreas segmentation and reveal the unsolved segmentation problems of the SOTA methods, such as the limited generalization ability on distinct medical centers, phases, and unseen diseases. To advance the unsolved problems, we further build four organ segmentation benchmarks for fully supervised, semi-supervised, weakly supervised, and continual learning, which are currently challenging and active research topics. Accordingly, we develop a simple and effective method for each benchmark, which can be used as out-of-the-box methods and strong baselines. We believe the AbdomenCT-1K dataset will promote future in-depth research towards clinical applicable abdominal organ segmentation methods.