Does non-COVID-19 lung lesion help? investigating transferability in COVID-19 CT image segmentation

doi:10.1016/j.cmpb.2021.106004

	Does non-COVID-19 lung lesion help? investigating transferability in COVID-19 CT image segmentation
	Wang, Yixin 1,2; Zhang, Yao 1,2; Liu, Yang 1,2; Tian, Jiang 2; Zhong, Cheng 2; Shi, Zhongchao 2; Zhang, Yang 1,3; He, Zhiqiang 1,3
	2021-04-01
发表期刊	COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
ISSN	0169-2607
卷号	202 页码:10
摘要	Background and Objective : Coronavirus disease 2019 (COVID-19) is a highly contagious virus spreading all around the world. Deep learning has been adopted as an effective technique to aid COVID-19 detection and segmentation from computed tomography (CT) images. The major challenge lies in the inadequate public COVID-19 datasets. Recently, transfer learning has become a widely used technique that leverages the knowledge gained while solving one problem and applying it to a different but related problem. However, it remains unclear whether various non-COVID19 lung lesions could contribute to segmenting COVID-19 infection areas and how to better conduct this transfer procedure. This paper provides a way to understand the transferability of non-COVID19 lung lesions and a better strategy to train a robust deep learning model for COVID-19 infection segmentation. Methods : Based on a publicly available COVID-19 CT dataset and three public non-COVID19 datasets, we evaluate four transfer learning methods using 3D U-Net as a standard encoder-decoder method. i) We introduce the multi-task learning method to get a multi-lesion pre-trained model for COVID-19 infection. ii) We propose and compare four transfer learning strategies with various performance gains and training time costs. Our proposed Hybrid encoder Learning strategy introduces a Dedicated-encoder and an Adapted-encoder to extract COVID-19 infection features and general lung lesion features, respectively. An attention-based Selective Fusion unit is designed for dynamic feature selection and aggregation. Results : Experiments show that trained with limited data, proposed Hybrid-encoder strategy based on multi-lesion pre-trained model achieves a mean DSC, NSD, Sensitivity, F1-score, Accuracy and MCC of 0.704, 0.735, 0.682, 0.707, 0.994 and 0.716, respectively, with better genetalization and lower over-fitting risks for segmenting COVID-19 infection. Conclusions : The results reveal the benefits of transferring knowledge from non-COVID19 lung lesions, and learning from multiple lung lesion datasets can extract more general features, leading to accurate and robust pre-trained models. We further show the capability of the encoder to learn feature representations of lung lesions, which improves segmentation accuracy and facilitates training convergence. In addition, our proposed Hybrid-encoder learning method incorporates transferred lung lesion features from nonCOVID19 datasets effectively and achieves significant improvement. These findings promote new insights into transfer learning for COVID-19 CT image segmentation, which can also be further generalized to other medical tasks. (c) 2021 Elsevier B.V. All rights reserved.
关键词	COVID-19 CT image Segmentation Transfer learning
DOI	10.1016/j.cmpb.2021.106004
收录类别	SCI
语种	英语
WOS研究方向	Computer Science ; Engineering ; Medical Informatics
WOS类目	Computer Science, Interdisciplinary Applications ; Computer Science, Theory & Methods ; Engineering, Biomedical ; Medical Informatics
WOS记录号	WOS:000639096300001
出版者	ELSEVIER IRELAND LTD
引用统计	被引频次：27[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/16670
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhang, Yang; He, Zhiqiang
作者单位	1.Chinese Acad Sci, Univ Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 2.Lenovo Res, AI Lab, Beijing, Peoples R China 3.Lenovo Ltd, Lenovo Corp Res & Dev, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Wang, Yixin,Zhang, Yao,Liu, Yang,et al. Does non-COVID-19 lung lesion help? investigating transferability in COVID-19 CT image segmentation[J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE,2021,202:10.
APA	Wang, Yixin.,Zhang, Yao.,Liu, Yang.,Tian, Jiang.,Zhong, Cheng.,...&He, Zhiqiang.(2021).Does non-COVID-19 lung lesion help? investigating transferability in COVID-19 CT image segmentation.COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE,202,10.
MLA	Wang, Yixin,et al."Does non-COVID-19 lung lesion help? investigating transferability in COVID-19 CT image segmentation".COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 202(2021):10.