浏览全部资源
扫码关注微信
收稿：2021-08-19，
修回：2021-11-5，
录用：2021-11-12，
纸质出版：2022-03-16
移动端阅览
目的
2
为辅助医生快速分辨新型冠状病毒肺炎(corona virus disease 2019
COVID-19)轻、重症患者，以便对症下药减轻医疗负担，提出一种基于结构图注意力网络的轻重症诊断算法。
方法
2
基于胸部CT图像提取的特定特征以及肺段间的位置关系构建结构图，以肺部内不同肺段为节点，以提取特征为节点属性。采用图神经网络汇聚相邻节点特征，再利用池化层获取分别代表左肺叶和右肺叶特征的图表示。使用结构注意力机制计算左、右肺叶的感染情况对结果诊断的重要性，并依据重要性融合左、右肺叶图表示以得到最终图表示，最后执行分类任务。由于数据中存在明显的类别不平衡现象，采用Focal-Loss损失函数优化模型以减轻对分类结果的影响。
结果
2
实验将所提算法分别与传统机器学习方法和流行的图神经网络算法做性能对比。在重症诊断的准确率上，本文算法相较于传统机器学习方法和图神经网络算法分别取得14.2%~42.0%和3.6%~4.8%的提升。在AUC(area under curve)指标上，本文算法相较于上述两种算法分别取得8.9%~18.7%和3.1%~3.6%的提升。除此之外，通过消融实验发现具有结构注意力机制的算法相较于未使用的算法在SPE(specificity)、SEN(sensitivity)和AUC 3个指标上分别取得了2.4%、1.4%和1.1%的提升；应用Focal-Loss损失函数的算法相较于未使用的算法提升了2.1%、1.1%和0.9%。
结论
2
所提出的诊断模型综合了图神经网络以及结构注意力机制的优点，引入Focal-Loss损失函数，提升了困难样本的分类准确率，使诊断结果更加准确。
Objective
2
The primary routine clinical diagnosis of COVID-19(corona virus disease 2019) is usually conducted based on epidemiological history
clinical manifestations and various laboratory detection methods
including nucleic acid amplification test (NAAT)
computed tomography (CT) scan and serological techniques. However
manual detection is costly
time-consuming and leads to the potential increase of the infection risk of clinicians. As a good alternative
artificial intelligence techniques on available data from laboratory tests play an important role in the confirmation of COVID-19 cases. Some studies have been designed for distinguishing between novel coronavirus pneumonia
community-acquired pneumonia and normal people by graph neural network. However
these studies leverage the relationships between features to build a topological structure graph (e.g.
connecting the nodes with high similarity)
while ignoring the inner relationships between different parts of the lung
and thus limiting the performance of their models. To address this issue
we propose a graph neural network with hierarchical information inherent to the physical structure of lungs for the improved diagnosis of COVID-19. Besides
an attention mechanism is introduced to capture the discriminative features of different severities of infection in the left and right lobes of different patients.
Method
2
Firstly
the topological structure is constructed based on the lungs' physical structure
and different lung segments are regarded as different nodes. Each node in the graph contains three kinds of handcrafted features
such as volume
density and mass feature
which reflect the infection in each lung segment and can be extracted from chest CT images using VB-Net. Secondly
based on graph neural network (GNN) and attention mechanism
we propose a novel structural attention graph neural network (SAGNN)
which can perform the graph classification task. The SAGNN first aggregates the features in a given sample graph
and then uses the attention mechanism to effectively fuse the different features to obtain the final graph representation. This representation is then fed into a linear layer with softmax activation function that performs graph classification
so that the corresponding sample graph can be finally classified as a mild case or a severe one. To alleviate the effect of category imbalance on the classification results
we use the focal loss function. We optimize the proposed model via back propagation and learn the representations of graphs.
Result
2
To verify the effectiveness of the proposed method
we compared SAGNN with several classical machine learning methods and graph classification methods on a real COVID-19 dataset
which includes 358 severe cases and 1 329 mild cases
provided by Shanghai Public Health Clinical Center. The result of comparative experiments was measured using three evaluation metrics including the sensitivity (SEN)
the specificity (SPE) and the area under the receiver operating characteristic(ROC) curve (AUC). In the experiments
our model had a good performance
indicating the effectiveness of our model. Based on comparison with the classical machine learning methods and the graph neural network methods
SAGNN outperformed by 14.2%~42.0% and 3.6%~4.8% in terms of SEN
respectively. In terms of AUC
the performance of SAGNN increased by 8.9%~18.7% and 3.1%~3.6%
respectively. In addition
through the ablation experiments of SAGNN
we found that the SAGNN with attention mechanism outperformed by 2.4%
1.4% and 1.1% in SPE
SEN and AUC than the SAGNN not with attention mechanism
respectively. The SAGNN with focal loss function outperformed by 2.1%
1.1% and 0.9% in SPE
SEN and AUC than the SAGNN with cross-entropy loss function
respectively.
Conclusion
2
In this work
we propose SAGNN
a new architecture for the diagnosis of severe and mild cases of COVID-19. Experimental results show the superior performance of SAGNN on classification task. Experimental results show that concatenating features of lung segments by their structure is effective. Moreover
we introduce an attention mechanism to distinguish the infection degree of right and left lungs. The focal loss is used to solve the issue of imbalanced group distribution
which further improves the overall network performance. We thus demonstrate the potential of SAGNN as clinical diagnosis support in this highly critical domain of medical intervention. We believe that our architecture provides a valuable case study for the early diagnosis of COVID-19
which is helpful for improvement in the field of computer-aided diagnosis and clinical practice.
Cao L J, Keerthi S S, Ong C J, Uvaraj V, Fu X J and Lee H P. 2006. Developing parallel sequential minimal optimization for fast training support vector machine. Neurocomputing, 70(1-3): 93-104
Cheung M and Moura J M F. 2020. Graph neural networks for COVID-19 drug discovery//Proceedings of 2020 IEEE International Conference on Big Data (Big Data). Atlanta, USA: IEEE: 5646-5648 [ DOI: 10.1109/bigdata50022.2020.9378164 http://dx.doi.org/10.1109/bigdata50022.2020.9378164 ]
Corman V M, Landt O, Kaiser M, Molenkamp R, Meijer A, Chu D K W, Bleicker T, Brünink S, Schneider J, Schmidt M L, Mulders D G J C, Haagmans B L, van der Veer B, van der Brink S, Wijsman L, Goderski G, Romette J L, Ellis J, Zambon M, Peiris M, Goossens H, Reusken C, Koopmans M P G and Drosten C. 2020. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Eurosurveillance, 25(3): #2000045 [DOI: 10.2807/1560-7917.es.2020.25.3.2000045]
Doshi S and Chepuri S P. 2020. Dr-COVID: graph neural networks for SARS-CoV-2 drug repurposing[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/2012.02151.pdf https://arxiv.org/pdf/2012.02151.pdf
Fritz C, Dorigatti E and Rügamer D. 2021. Combining graph neural networks and spatio-temporal disease models to predict COVID-19 cases in Germany[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/2101.00661.pdf https://arxiv.org/pdf/2101.00661.pdf
Goncharov M, Pisov M, Shevtsov A, Shirokikh B, Kurmukov A, Blokhin I, Chernina V, Solovev A, Gombolevskiy V, Morozov S and Belyaev M. 2021. CT-based COVID-19 triage: deep multitask learning improves joint identification and severity quantification. Medical Image Analysis, 71: #102054 [DOI: 10.1016/j.media.2021.102054]
Hamilton W L, Ying R and Leskovec J. 2017. Inductive representation learning on large graphs[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/1706.02216.pdf https://arxiv.org/pdf/1706.02216.pdf
Ioannidis V N, Zheng D and Karypis G. 2020. Few-shot link prediction via graph neural networks for COVID-19 drug-repurposing[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/2007.10261.pdf https://arxiv.org/pdf/2007.10261.pdf
Kapoor A, Ben X, Liu L Y, Perozzi B, Barnes M, Blais M and O'Banion S. 2020. Examining COVID-19 forecasting using spatio-temporal graph neural networks[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/2007.03113.pdf https://arxiv.org/pdf/2007.03113.pdf
Kipf T N and Welling M. 2016. Semi-supervised classification with graph convolutional networks[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/1609.02907.pdf https://arxiv.org/pdf/1609.02907.pdf
Micheli A. 2009. Neural network for graphs: a contextual constructive approach. IEEE Transactions on Neural Networks, 20(3): 498-511 [DOI: 10.1109/tnn.2008.2010350]
Ng A Y and Jordan M I. 2000. On discriminative vs. generative classifiers: A comparison of logistic regression and naive Bayes// Proceedings of the 14th International Conference on Neural Information Processing Systems: Natural and Synthetic. Colorado, USA: [s. n.]: 841-848
Rakhimberdina Z and Murata T. 2019. Linear graph convolutional model for diagnosing brain disorders//Proceedings of the 8th International Conference on Complex Networks and Their Applications. Lisbon, Portugal: Springer: 815-826 [ DOI: 10.1007/978-3-030-36683-4_65 http://dx.doi.org/10.1007/978-3-030-36683-4_65 ]
Saha P, Mukherjee D, Singh P K, Ahmadian A, Ferrara M and Sarkar R. 2021. GraphCOVIDnet: a graph neural network based model for detecting COVID-19 from CT scans and X-rays of chest. Scientific Reports, 11(1): #8304 [DOI: 10.1038/s41598-021-87523-1]
Scarselli F, Gori M, Tsoi A C, Hagenbuchner M and Monfardini G. 2009. The graph neural network model. IEEE Transactions on Neural Networks, 20(1): 61-80 [DOI: 10.1109/tnn.2008.2005605]
Shan F, Gao Y Z, Wang J, Shi W Y, Shi N N, Han M F, Xue Z, Shen D G and Shi Y X. 2020. Lung infection quantification of COVID-19 in CT images with deep learning[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/2003.04655.pdf https://arxiv.org/pdf/2003.04655.pdf
Tian Z Q, Li X J, Zheng Y Y, Chen Z, Shi Z, Liu L Z and Fei B W. 2020. Graph-convolutional-network-based interactive prostate segmentation in MR images. Medical Physics, 47(9): 4164-4176 [DOI: 10.1002/mp.14327]
Van Der Maaten L. 2014. Accelerating t-SNE using tree-based algorithms. The Journal of Machine Learning Research, 15(1): 3221-3245
Wang S H, Govindaraj V V, Górriz J M, Zhang X and Zhang Y D. 2021. COVID-19 classification by FGCNet with deep feature fusion from graph convolutional network and convolutional neural network. Information Fusion, 67: 208-229 [DOI: 10.1016/j.inffus.2020.10.004]
Wei W, Hu X W, Cheng Q, Zhao Y M and Ge Y Q. 2020. Identification of common and severe COVID-19: the value of CT texture analysis and correlation with clinical characteristics. European Radiology, 30(12): 6788-6796 [DOI: 10.1007/s00330-020-07012-3]
World Health Organization. 2020. Report of the WHO-China joint mission oncoronavirus disease 2019 (COVID-19)[EB/OL]. [2021-07-15] . https://www.who.int/publications/i/item/report-of-the-who-china-joint-mission-on-coronavirus-disease-2019-(covid-19) https://www.who.int/publications/i/item/report-of-the-who-china-joint-mission-on-coronavirus-disease-2019-(covid-19)
Wu Z Y and McGoogan J M. 2020. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72 314 cases from the Chinese Center for Disease Control and Prevention. JAMA, 323(13): 1239-1242 [DOI: 10.1001/jama.2020.2648]
Xue J H and Titterington D M. 2002. Comment on "on discriminative vs. generative classifiers: a comparison of logistic regression and naïve Bayes". Neural Processing Letters, 28(3): 169-187
Xu K, Hu W H, Leskovec J and Jegelka S. 2018. How powerful are graph neural networks?[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/1810.00826.pdf https://arxiv.org/pdf/1810.00826.pdf
Ying R, You J, Morris C, Ren X, Hamilton W L and Leskovec J. 2018. Hierarchical graph representation learning with differentiable pooling[EB/OL]. [2021-07-15] . https://arxiv.org/pdf/1806.08804.pdf https://arxiv.org/pdf/1806.08804.pdf
Zhang M H and Chen Y X. 2018. Link prediction based on graph neural networks [EB/OL]. [2021-07-15] . Proceedings of Advances in Neural Information Processing Systems 31: Annual Conference on Neural Information Processing Systems 2018. Montréal, Canada: [s. n.]: 5171-5181. https://arxiv.org/pdf/1806.08804.pdf https://arxiv.org/pdf/1806.08804.pdf
Zhang M H, Cui Z C, Neumann M and Chen Y X. 2018. An end-to-end deep learning architecture for graph classification//Proceedings of the 32nd AAAI Conference on Artificial Intelligence. New Orleans, USA: AAAI: 4438-4445
Zhu X F, Song B, Shi F, Chen Y B, Hu R Y, Gan J Z, Zhang W H, Li M, Wang L Y, Gao Y Z, Shan F and Shi D G. 2021. Joint prediction and time estimation of COVID-19 developing severe symptoms using chest CT scan. Medical Image Analysis, 67: #101824 [DOI: 10.1016/j.media.2020.101824]
Zuo Y, Huang G and Nie S D. 2021. Application and challenges of deep learning in the intelligent processing of medical images. Journal of Image and Graphics, 26(2): 305-315
左艳, 黄钢, 聂生东. 2021. 深度学习在医学影像智能处理中的应用与挑战. 中国图象图形学报, 26(2): 305-315[DOI: 10.11834/jig.190470]
相关作者
相关机构
京公网安备11010802024621
