基于注意力机制和Inf-Net的新冠肺炎图像分割方法

doi:10.16180/j.cnki.issn1007-7820.2023.02.004

摘要/Abstract

摘要：

新型冠状病毒肺炎肆虐全球,严重影响了人类社会的生活和健康。CT影像技术是检测新冠肺炎的重要诊断方式,从CT图像中自动准确分割出新冠肺炎病灶区域,对于诊断、治疗和预后都有重要意义。针对新冠肺炎病灶的自动分割,文中提出基于Inf-Net算法改进的自动分割方法,通过引入通道注意力机制加强特征表示,并运用注意力门模块来更好地融合边缘信息。在COVID-19 CT分割数据集上的实验结果表明,文中所提出新冠肺炎图像分割方法的Dice系数、灵敏度、特异率分别为75.1%、75.4%和95.4%,算法性能也优于部分主流方法。

关键词: 新冠肺炎, 计算机断层扫描影像, 医学图像分割, 语义分割, 注意力机制, Inf-Net网络, 卷积神经网络, 深度学习

Abstract:

COVID-19 pandemic has recently ravaged the world, seriously affecting the life and health of human society. CT imaging technology is an important diagnostic method for detecting COVID-19. Automatic and accurate segmentation of the lesion is of great significance for diagnosis, treatment and prognosis. In view of the segmentation of new coronary pneumonia lesions, an improved automatic segmentation method based on the Inf-Net algorithm is proposed, which introduces the channel attention module to improve feature representation and attention gate model to better fuse edge information. The experimental results on COVID-19 CT segmentation datasets show that the Dice similarity coefficient, sensitivity and specificity of the proposed method are 75.1%, 75.4% and 95.4%, respectively, and the algorithm performance is also better than some mainstream methods.

Key words: COVID-19, computerized tomography image, medical image segmentation, semantic segmentation, attention mechanism, Inf-Net network, convolutional neural network, deep learning

中图分类号:

TP391

左斌,李菲菲. 基于注意力机制和Inf-Net的新冠肺炎图像分割方法[J]. 电子科技, 2023, 36(2): 22-28.

ZUO Bin,LI Feifei. An Effective Segmentation Method for COVID-19 CT Image Based on Attention Mechanism and Inf-Net[J]. Electronic Science and Technology, 2023, 36(2): 22-28.

图/表 9

图1

图2

图3

图4

图5

图6

表1

实验结果比较分析"

网络	Dice	灵敏度	特异率	S_α	$E φ m e a n$	MAE
U-Net^[14]	0.439	0.534	0.858	0.622	0.625	0.186
Attention U-Net^[7]	0.583	0.637	0.921	0.744	0.739	0.112
Gated-UNet^[23]	0.623	0.658	0.926	0.725	0.814	0.102
Dense-UNet^[24]	0.515	0.594	0.840	0.655	0.662	0.184
U-Net++^[15]	0.581	0.672	0.902	0.722	0.720	0.120
Inf-Net^[6]	0.682	0.692	0.943	0.781	0.838	0.082
mInf-Net (Ours)	0.690	0.740	0.923	0.771	0.842	0.092
Semi-mInf-Net (Ours)	0.751	0.754	0.954	0.799	0.904	0.064

表1

图7

表2

消融实验结果"

网络	Dice	灵敏度	特异率	S_α	$E φ m e a n$	MAE
Backbone	0.739	0.725	0.960	0.800	0.894	0.064
Backbone+CAM	0.742	0.727	0.959	0.790	0.897	0.064
Backbone+AG	0.750	0.758	0.953	0.798	0.904	0.064
Backbone+CAM+AG	0.751	0.754	0.954	0.799	0.904	0.064

表2

参考文献 24

[1]	Dong E, Du H, Gardner L. An interactive web-based dashboard to track COVID-19 in real time[J]. The Lancet Infectious Diseases, 2020, 20(5):533-534. doi: 10.1016/S1473-3099(20)30120-1
[2]	缪冉, 李菲菲, 陈虬. 基于卷积神经网络与多尺度空间编码的场景识别方法[J]. 电子科技, 2020, 33(12):54-58.
	Miao Ran, Li Feifei, Chen Qiu. Scene recognition algorithm based on convolutional neural networks and multi-scale space encoding[J]. Electronic Science and Technology, 2020, 33(12):54-58.
[3]	Ma Y, Feng P, He P, et al. Segmenting lung lesions of COVID-19 from CT images via pyramid pooling improved Unet[J]. Biomedical Physics & Engineering Express, 2021, 7(4):1-12.
[4]	周子棋, 康莉, 黄建军. 多站点新冠肺炎肺部CT图像的三维深度卷积分割[J]. 信号处理, 2021, 35(5):771-779.
	Zhou Ziqi, Kang Li, Huang Jianjun. 3D CNN segmentation of multi-site lung CT COVID-19 lesion[J]. Journal of Signal Processing, 2021, 35(5):771-779.
[5]	宋瑶, 刘俊. 改进U-Net的新冠肺炎图像分割方法[J]. 计算机工程与应用, 2021, 57(19):243-251. doi: 10.3778/j.issn.1002-8331.2010-0207
	Song Yao, Liu Jun. Improved U-Net network for COVID-19 image segmentation[J]. Computer Engineering and Applications, 2021, 57(19):243-251. doi: 10.3778/j.issn.1002-8331.2010-0207
[6]	Fan D P, Zhou T, Ji G P, et al. Inf-Net: Automatic COVID-19 lung infection segmentation from CT images[J]. IEEE Transactions on Medical Imaging, 2020, 39(8):2626-2637. doi: 10.1109/TMI.2020.2996645
[7]	Oktay O, Schlemper J, Folgoc L L, et al. Attention U-Net: Learning where to look for the pancreas[C]. Amsterdam: Proceedings of the Medical Imaging with Deep Learning, 2018.
[8]	Zhao J X, Liu J J, Fan D P, et al. EGNet: Edge guidance network for salient object detection[C]. Seoul: IEEE International Conference on Computer Vision, 2019.
[9]	Wu Z, Su L, Huang Q. Stacked cross refinement network for edge-aware salient object detection[C]. Seoul: IEEE International Conference on Computer Vision, 2019.
[10]	Zhang Z, Fu H, Dai H, et al. ET-Net: A generic edge-attention guidance network for medical image segmentation[C]. Shenzhen: International Conference on Medical Image Computing and Computer Assisted Intervention, 2019.
[11]	Fu H, Cheng J, Xu Y, et al. Joint optic disc and cup segmentation based on multi-label deep network and polar transformation[J]. IEEE Transactions on Medical Imaging, 2018, 37(7):1597-1605. doi: 10.1109/TMI.2018.2791488 pmid: 29969410
[12]	Gu Z, Cheng J, Fu H, et al. CE-Net: Context encoder network for 2D medical image segmentation[J]. IEEE Transactions on Medical Imaging, 2019, 38(10):2281-2292. doi: 10.1109/TMI.2019.2903562 pmid: 30843824
[13]	Zhang S, Fu H, Yan Y, et al. Attention guided network for retinal image segmentation[C]. Shenzhen: International Conference on Medical Image Computing and Computer Assisted Intervention, 2019.
[14]	Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation[C]. Munich: International Conference on Medical Image Computing and Computer Assisted Intervention, 2015.
[15]	Zhou Z, Siddiquee M M R, Tajbakhsh N, et al. UNet++: A nested U-Net architecture for medical image segmentation[J]. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, 2018, 11045:3-11. doi: 10.1007/978-3-030-00889-5_1 pmid: 32613207
[16]	Gao S H, Cheng M M, Zhao K, et al. Res2Net: A new multi-scale backbone architecture[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 43(2):652-662. doi: 10.1109/TPAMI.2019.2938758
[17]	Wu Z, Su L, Huang Q. Cascaded partial decoder for fast and accurate salient object detection[C]. Long Beach: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019.
[18]	Fu J, Liu J, Tian H, et al. Dual attention network for scene segmentation[C]. Long Beach: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019.
[19]	Wei Y, Feng J, Liang X, et al. Object region mining with adversarial erasing: A simple classifcation to semantic segmentation approach[C]. Honolulu: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017.
[20]	Chen S, Tan X, Wang B, et al. Reverse attention-based residual network for salient object detection[J]. IEEE Transactions on Image Processing, 2020, 29(2):3763-3776. doi: 10.1109/TIP.2020.2965989
[21]	Fan D P, Cheng M M, Liu Y, et al. Structure-measure: A new way to evaluate foreground maps[C]. Venice: Proceedings of the IEEE International Conference on Computer Vision, 2017.
[22]	Fan D P, Gong C, Cao Y, et al. Enhanced-alignment measure for binary foreground map evaluation[C]. Stockholm: International Joint Conferences on Artificial Intelligence, 2018.
[23]	Schlemper J, Oktay O, Schaap M, et al. Attention gated networks: Learning to leverage salient regions in medical images[J]. Medical Image Analysis, 2019, 53(6):197-207. doi: 10.1016/j.media.2019.01.012
[24]	Li X, Chen H, Qi X, et al. H-DenseUNet: Hybrid densely connected UNet for liver and tumor segmentation from CT volumes[J]. IEEE Transactions on Medical Imaging, 2018, 37(12):2663-2674. doi: 10.1109/TMI.2018.2845918 pmid: 29994201