Robustness study of noisy annotation in deep learning based medical image segmentation.
Robustness study of noisy annotation in deep learning based medical image segmentation.
复制标题
DOI：
10.1088/1361-6560/ab99e5
复制
发表时间：
2020-08-27
影响因子：
3.5
通讯作者：
Lu W
中科院分区：
文献类型：
--
作者：
Yu S;Chen M;Zhang E;Wu J;Yu H;Yang Z;Ma L;Gu X;Lu W
部分由于使用了详尽的注释数据，深度网络在医学图像分割方面取得了令人印象深刻的性能。然而，与噪声注释配对的医学成像数据是普遍存在的，但关于噪声注释对基于深度学习的医学图像分割的影响知之甚少。我们研究了噪声注释的上下文中的下颌骨分割从CT图像的效果。首先，从我们的临床数据库中收集了202例头颈部癌症患者的图像，其中危险器官由12名计划剂量师之一注释。下颌骨被粗略地诠释为规划回避结构。然后，由头颈部专家检查并校正下颌骨标签，以获得参考标准。最后，通过改变训练集中噪声标签的比例，对深度网络进行训练和测试，用于下颌骨分割。训练后的模型在另外两个公共数据集上进行了进一步测试。实验结果表明，使用噪声标签训练的网络比使用参考标准训练的网络具有更差的分割，并且通常，噪声标签越少，性能越好。当使用20%或更少的噪声情况进行训练时，通过噪声或参考注释训练的模型之间的分割结果没有显著差异。交叉数据集验证结果验证了用噪声数据训练的模型与用参考标准训练的模型相比具有竞争力的性能。该研究表明，所涉及的网络是鲁棒的噪声注释在一定程度上从CT图像的下颌骨分割。它还强调了标签质量在深度学习中的重要性。在未来的工作中，应该特别注意如何利用少量的参考标准样本来提高带噪声注释的深度学习的性能。
Partly due to the use of exhaustive-annotated data, deep networks have achieved impressive performance on medical image segmentation. Medical imaging data paired with noisy annotation are, however, ubiquitous, but little is known about the effect of noisy annotation on deep learning based medical image segmentation. We studied the effect of noisy annotation in the context of mandible segmentation from CT images. First, 202 images of head and neck cancer patients were collected from our clinical database, where the organs-at-risk were annotated by one of twelve planning dosimetrists. The mandibles were roughly annotated as the planning avoiding structure. Then, mandible labels were checked and corrected by a head and neck specialist to get the reference standard. At last, by varying the ratios of noisy labels in the training set, deep networks were trained and tested for mandible segmentation. The trained models were further tested on other two public datasets. Experimental results indicated that the network trained with noisy labels had worse segmentation than that trained with reference standard, and in general, fewer noisy labels led to better performance. When using 20% or less noisy cases for training, no significant difference was found on the segmentation results between the models trained by noisy or reference annotation. Cross-dataset validation results verified that the models trained with noisy data achieved competitive performance to that trained with reference standard. This study suggests that the involved network is robust to noisy annotation to some extent in mandible segmentation from CT images. It also highlights the importance of labeling quality in deep learning. In the future work, extra attention should be paid to how to utilize a small number of reference standard samples to improve the performance of deep learning with noisy annotation.
登录
查看更多内容
影响因子：
4.4
作者：
Clark, Kenneth;Vendt, Bruce;Prior, Fred
通讯作者：
Prior, Fred
NiftyNet：医学成像深度学习平台。
影响因子：
6.1
作者：
Gibson E;Li W;Sudre C;Fidon L;Shakir DI;Wang G;Eaton-Rosen Z;Gray R;Doel T;Hu Y;Whyntie T;Nachev P;Modat M;Barratt DC;Ourselin S;Cardoso MJ;Vercauteren T
通讯作者：
Vercauteren T
影响因子：
3.7
作者：
Liu Y;Stojadinovic S;Hrycushko B;Wardak Z;Lau S;Lu W;Yan Y;Jiang SB;Zhen X;Timmerman R;Nedzi L;Gu X
通讯作者：
Gu X
影响因子：
3.8
作者：
Raudaschl, Patrik F.;Zaffino, Paolo;Fritscher, Karl D.
通讯作者：
Fritscher, Karl D.
影响因子：
3.5
作者：
Chen, Haibin;Lu, Weiguo;Gu, Xuejun
通讯作者：
Gu, Xuejun