迁移学习和注意机制融合用于CT图像COVID-19病灶分割的计算机辅助诊断

doi:10.1007/s12204-023-2646-3

摘要/Abstract

摘要： 肺部疾病的类型多样且复杂，基于计算机断层扫描技术（Computed Tomography，CT）图像进行高质量的肺病变区域分割成为计算机辅助诊断的重难点问题。研究将注意力机制与迁移学习相结合，构建可在肺CT图像上实现新冠肺炎自动诊断的深度学习模型。迁移学习中编码阶段使用ImageNet预先训练的VGG16作为编码器，基于U-Net结构设计解码器，在每次进行拼接操作时引入注意力模块能让模型对重要信息重点关注并快速锁定重要部分。在COVID-19-CT-Seg-Benchmark数据集上进行实验，Dice，F1，Accuracy最高分别达到了0.9071，0.9076，0.9965。同时评估泛化性能，性能指标包括Dice，F1和Accuracy超过0.8。实验结果表明本文所提出的分割网络的有效性。

关键词: 迁移学习, 注意力机制, 计算机断层扫描成像, COVID-19, 图像分割

Abstract: Various and intricate varieties of lung disease have made it challenging for computer aided diagnosis to appropriately segment lung lesions utilizing computed tomography (CT) images. This study integrates transfer learning with the attention mechanism to construct a deep learning model that can automatically detect new coronary pneumonia on lung CT images. In this study, using VGG16 pre-trained by ImageNet as the encoder, the decoder was established utilizing the U-Net structure. The attention module is incorporated during each concatenate procedure, permitting the model to concentrate on the critical information and identify the crucial components efficiently. The public COVID-19-CT-Seg-Benchmark dataset was utilized for experiments, and the highest scores for Dice, F1, and Accuracy were 0.907 1, 0.907 6, and 0.996 5, respectively. The generalization performance was assessed concurrently, with performance metrics including Dice, F1, and Accuracy over 0.8. The experimental findings indicate the feasibility of the segmentation network proposed in this study.

中图分类号:

TP391
R563

. 迁移学习和注意机制融合用于CT图像COVID-19病灶分割的计算机辅助诊断[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 566-581.

Fan Xinggang, Liu Jiaxian, Li Chao, Yang Youdong, Gu Wenting, Jiang Xinyang. Computer Aided Diagnosis for COVID-19 in CT Images Utilizing Transfer Learning and Attention Mechanism[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 566-581.

参考文献

[1] ISLAMI F, WARD E M, SUNG H, et al. Annual report to the nation on the status of cancer, part 1: national cancer statistics[J]. JNCI: Journal of the National Cancer Institute, 2021, 113(12): 1648-1669.

[2] CRESSMAN S, PEACOCK S J, TAMMEMÄGI M C, et al. The cost-effectiveness of high-risk lung cancer screening and drivers of program efficiency [J]. Journal of Thoracic Oncology, 2017, 12(8): 1210-1222.

[3] FIELD J K, VULKAN D, DAVIES M P A, et al. Lung cancer mortality reduction by LDCT screening: UKLS randomised trial results and international meta-analysis [J]. The Lancet Regional Health Europe, 2021, 10: 100179.

[4] OKTAY O, SCHLEMPER J, LE FOLGOC L, et al. Attention U-net: Learning where to look for the pancreas [DB/OL]. (2018-04-11). https://arxiv.org/abs/1804.03999

[5] SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4): 640-651.

[6] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks [J]. Communications of the ACM, 2017, 60(6): 84-90.

[7] RONNEBERGER O, FISCHER P, BROX T. U-net: Convolutional networks for biomedical image segmentation[M]// Medical image computing and computer-assisted intervention – MICCAI 2015. Cham: Springer, 2015: 234-241.

[8] PAN S J, YANG Q. A survey on transfer learning [J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359.

[9] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[M]// Computer vision – ECCV 2018. Cham: Springer, 2018: 3-19.

[10] BADRINARAYANAN V, KENDALL A, CIPOLLA R. SegNet: A deep convolutional encoder-decoder architecture for image segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.

[11] SIDDIQUE N, PAHEDING S, ELKIN C P, et al. U-net and its variants for medical image segmentation: A review of theory and applications [J]. IEEE Access, 2021, 9: 82031-82057.

[12] CHENG D, LAM E Y. Transfer learning U-net deep learning for lung ultrasound segmentation [DB/OL]. (2021-10-05). https://arxiv.org/abs/2110.02196

[13] YIN J T, LI J W, HUANG Q H, et al. Ultrasonographic segmentation of fetal lung with deep learning [J]. Journal of Biosciences and Medicines, 2021, 9(1): 146-153.

[14] PETIT O, THOME N, RAMBOUR C, et al. U-net transformer: Self and cross attention for medical image segmentation[M]// Machine learning in medical imaging. Cham: Springer, 2021: 267-276.

[15] HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023.

[16] JADERBERG M, SIMONYAN K, ZISSERMAN A, et al. Spatial transformer networks [DB/OL]. (2015-06-05). https://arxiv.org/abs/1506.02025

[17] AMIRI M, BROOKS R, RIVAZ H. Fine-tuning U-net for ultrasound image segmentation: Different layers, different outcomes [J]. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2020, 67(12): 2510-2518.

[18] TAJBAKHSH N, SHIN J Y, GURUDU S R, et al. Convolutional neural networks for medical image analysis: Full training or fine tuning? [J]. IEEE Transactions on Medical Imaging, 2016, 35(5): 1299-1312.

[19] PRAKASH R M, THENMOEZHI N, GAYATHRI M. Face recognition with convolutional neural network and transfer learning [C]//2019 International Conference on Smart Systems and Inventive Technology. Tirunelveli: IEEE, 2019: 861-864.

[20] MA J, WANG Y X, AN X L, et al. Toward data-efficient learning: A benchmark for COVID-19 CT lung and infection segmentation [J]. Medical Physics, 2021, 48(3): 1197-1210.

[21] XIA C Q, LI J, CHEN X W, et al. What is and What is Not a Salient Object? Learning Salient Object Detector by Ensembling Linear Exemplar Regressors [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 4399-4407.

[22] XIE Y T, RICHMOND D. Pre-training on grayscale ImageNet improves medical image classification[M]// Computer vision – ECCV 2018 workshops. Cham: Springer, 2019: 476-484.

[23] ZOU K H, WARFIELD S K, BHARATHA A, et al. Statistical validation of image segmentation quality based on a spatial overlap index 1 [J]. Academic Radiology, 2004, 11(2): 178-189.

[24] IANDOLA F N, HAN S, MOSKEWICZ M W, et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size [DB/OL]. (2016-02-24). https://arxiv.org/abs/1602.07360

[25] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778.

[26] JOSEPH RAJ A N, ZHU H P, KHAN A, et al. ADID-UNET—a segmentation model for COVID-19 infection from lung CT scans [J]. PeerJ Computer Science, 2021, 7: e349.