为了解决肺部病灶图像对比度低、纹理细节模糊以及边缘特征提取不充分等难题，本文提出一种融合双重注意力机制的轻量级U型网络。首先，设计一种注意双分支组合模块。在编码阶段，两分支分别关注全局与局部信息，获取病灶全局定位信息与边缘特征；其次，同时引入并行的纹理增强模块，使用量化计数算子得到统计特征直方图，增强浅层网络提取的纹理特征，解决了对比度低的挑战；最后，构建反注意双干扰细分模块，在解码阶段使网络关注并处理误分割信息，实现重构图像内假阳性与假阴性特征的同时消除。在肺部病灶数据集COVID-19 CT scan和MS COVID-19上验证了网络的有效性。与现有网络相比，本文所提网络在5个评价指标上均取得了最优结果，Dice指标相比次优的UNeXt模型提高了1.42%，且同时实现了参数轻量化的效果。
To address the problems of low contrast, fuzzy texture details, and inadequate edge feature extraction in lung lesion images, this paper proposed a novel lightweight U-shaped network incorporating dual attention, termed DALU-Net. First, an attentionbased two-branch fusion module was designed. During encoding, the two branches focused on global and local information, respectively, to capture global localization information and lesion edge features. Then, a parallel texture enhancement module was introduced,and a statistical feature histogram was obtained using a quantization counting operator to enhance the texture features extracted by the shallow network and address the challenge of low contrast. Finally, a reverse-attention dual-interference refinement module was developed to enable the network to focus on and process mis-segmentation information during the decoding stage, thus simultaneously eliminating false-positive and false-negative features in the reconstructed image. The effectiveness of the network was verified on two lung lesion datasets: COVID-19 CT scan and MS COVID-19. Compared with existing networks,the proposed network achieves the best results on all five evaluation metrics, with a Dice score that is 1.42% higher than that of the second-best model, UNeXt, while also using fewer parameters.
[1] Wang D, Zhang J, Lau J, et al. Mechanisms of lung disease development in rheumatoid arthritis [J]. Nature Reviews Rheumatology, 2019, 15(10): 581-596.
[2] Zhang X M, Zhou B R, Osborn T, et al. Lung ultrasound surface wave elastography for assessing interstitial lung disease [DB/OL]. (2018-06-09) [2023-09-15]. https://arxiv.org/abs/1806.03500.
[3] Pu J T, Paik D S, Meng X, et al. Shape “break-and-repair” strategy and its application to automated medical image segmentation [J]. IEEE Transactions on Visualization and Computer Graphics, 2011, 17(1): 115-124.
[4] Xu R, Pan J, Ye X C, et al. A pilot study to utilize a deep convolutional network to segment lungs with complex opacities [C]//Chinese Automation Congress, 2017: 3291-3295.
[5] Alves J H, Neto P M M, Oliveira L F. Extracting lungs from CT images using fully convolutional networks [C]//International Joint Conference on Neural Networks, 2018: 1-8.
[6] Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation [C]//IEEE Conference on Computer Vision and Pattern Recognition, 2015: 3431-3440.
[7] Sabour S, Frosst N, Hinton G E. Dynamic routing between capsules [DB/OL]. (2017-11-07) [2023-09-15]. https://arxiv.org/abs/1710.09829.
[8] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation [C]//18th International Conference on Medical Image Computing and ComputerAssisted Intervention (MICCAI), 2015: 234-241.
[9] Ma Y L, Zhu Z B, Dong Z K, et al. Multichannel retinal blood vessel segmentation based on the combination of matched filter and U-Net network [J]. BioMed Research International, 2021, 2021: 5561125.
[10] Diakogiannis F I, Waldner F, Caccetta P, et al. ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 162: 94-114.
[11] Choi H S, Kim J H, Huh J, et al. Phase-aware speech enhancement with deep complex U-Net [DB/OL]. (2019-04-02) [2023-09-15]. https://arxiv.org/abs/1903.03107.
[12] 高红霞, 郜伟. 融合密集连接与自适应加权损失的血管壁图像分割[J]. 计算机应用研究, 2022, 39(6): 1905-1910. Gao H X, Gao W. Vessel wall image segmentation based on dense connection and adaptive weighted loss [J]. Computer Application Research, 2022, 39(6): 1905-1910. (in Chinese)
[13] Wang B, Wei W, Qiu S, et al. Boundary aware U-Net for retinal layers segmentation in optical coherence tomography images [J]. IEEE Journal of Biomedical and Health Informatics, 2021, 25(8): 3029-3040.
[14] 郭薇, 何聪, 宫照煊, 等. 基于W型网络的乳腺动态对比度增强磁共振图像合成[J]. 计算机应用研究, 2022, 39(1): 312-315. Guo W, He C, Gong Z X, et al. Dynamic contrast-enhanced magnetic resonance imaging synthesis of breast using W-net [J]. Application Research of Computers, 2022, 39(1): 312-315. (in Chinese)
[15] Qiu Y, Liu Y, Li S J, et al. MiniSeg: an extremely minimum network for efficient COVID-19 segmentation [C]//35th AAAI Conference on Artificial Intelligence/33rd Conference on Innovative Applications of Artificial Intelligence/11th Symposium on Educational Advances in Artificial Intelligence, 2021: 4846-4854.
[16] Wang Q L, Wu B G, Zhu P F, et al. ECA-Net: efficient channel attention for deep convolutional neural networks [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11531-11539.
[17] Hou Q B, Zhou D Q, Feng J S. Coordinate attention for efficient mobile network design [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 13708-13717.
[18] Wu C, Long C, Li S J, et al. MSRAformer: multiscale spatial reverse attention network for polyp segmentation [J]. Computers in Biology and Medicine, 2022, 151: 106274.
[19] Zhu L Y, Ji D Y, Zhu S P, et al. Learning statistical texture for semantic segmentation [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021: 12532-12541.
[20] Ma J, Ge C, Wang Y X, et al. COVID-19 CT lung and infection segmentation dataset [EB/OL]. (2020-04-20) [2023-09-15]. https://zenodo.org/records/3757476.
[21] Wang H Y, Xie S A, Lin L F, et al. Mixed transformer U-Net for medical image segmentation [DB/OL]. (2021-11-11) [2023-09-15]. https://arxiv.org/abs/2111.04734.
[22] Valanarasu J M J, Patel V M. UNeXt: MLP-based rapid medical image segmentation network [C]//25th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2022: 23-33.
[23] Wang H N, Cao P, Liu X L, et al. Narrowing the semantic gaps in U-Net with learnable skip connections: the case of medical image segmentation [J]. Neural Networks, 2024, 178: 106546.
[24] Dong B, Wang P C, Wang F. Head-free lightweight semantic segmentation with linear transformer [C]//37th AAAI Conference on Artificial Intelligence (AAAI)/35th Conference on Innovative Applications of Artificial Intelligence/13th Symposium on Educational Advances in Artificial Intelligence, 2023: 516-524.
[25] Dinh B D, Nguyen T T, Tran T T, et al. 1M parameters are enough? A lightweight CNN-based model for medical image segmentation [C]//Asia-Pacific-Signal-and-Information-Processing-Association Annual Summit and Conference (APSIPA ASC), 2023: 1279-1284.
