广东工业大学学报 ›› 2025, Vol. 42 ›› Issue (1): 24-32.doi: 10.12052/gdutxb.230177

• 智慧医疗 • 上一篇    

基于Transformer与注意力机制的肺部肿瘤分割方法

曾安1, 王丹1, 杨宝瑶1, 张小波2, 石镇维3, 刘再毅3, 潘丹4   

  1. 1. 广东工业大学 计算机学院, 广东 广州 510006;
    2. 广东工业大学 自动化学院, 广东 广州 510006;
    3. 广东省人民医院, 广东 广州 510080;
    4. 广东技术师范大学 电子与信息学院, 广东 广州 510665
  • 收稿日期:2023-11-11 发布日期:2025-01-06
  • 作者简介:曾安(1978–)，女，教授，博士生导师，主要研究方向为图像处理、模式识别、人工智能，E-mail：zengan@gdut.edu.cn
  • 基金资助:
    广东省科技计划项目(2019A050510041) ；国家自然科学基金资助项目(61976058，6210209，62102098) ；广东省重点领域研发计划项目(2021B0101220006) ；广东省自然科学基金资助项目(2021A1515012300，2022A1515140096) ；广州市科技计划项目(202103000034，202206010007，202201010266) ；云南省重大科技专项(202102AA100012)

Lung Tumor Segmentation Method Based on Transformer and Attention Mechanisms

Zeng An1, Wang Dan1, Yang Baoyao1, Zhang Xiaobo2, Shi Zhenwei3, Liu Zaiyi3, Pan Dan4   

  1. 1. School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, China;
    2. School of Automation, Guangdong University of Technology, Guangzhou 510006, China;
    3. Guangdong Provincial People's Hospital, Guangzhou 510080, China;
    4. School of Electronics and Information, Guangdong Polytechnic Normal University, Guangzhou 510665, China
  • Received:2023-11-11 Published:2025-01-06

HTML

50

PDF (PC)

317

摘要： 肺部肿瘤的准确分割对于肿瘤的诊断和治疗具有重要作用，然而肺部肿瘤分割中存在病灶与周围组织的对比度低、肿瘤与正常组织易粘连和背景噪声大等问题。针对这些问题，本文提出了一种基于Transformer和注意力机制的肺部肿瘤分割方法。在Transformer编码器阶段引入全局和局部的注意力机制，使得网络可以同时关注全局和局部的上下文信息；在跳跃连接阶段，使用通道优先卷积注意力机制，可以增强复杂病灶的空间感知能力和降低通道维度冗余，从而提高肿瘤的分割精度。在私有数据集GDPH和公共数据集LUNG1上的测试结果表明，本文方法相比其他8种分割方法，Dice指标在两个数据集上表现最优，分别为90.96%和88.18%，可以为临床的诊疗提供可靠辅助。

关键词: 肺部肿瘤, 医学图像分割, 卷积神经网络, Transformer, 注意力机制

Abstract: The accurate segmentation of lung tumors plays a crucial role in tumor diagnosis and treatment. However, lung tumor segmentation is often challenged by several issues such as low contrast between lesions and surrounding tissues, tumor-normal tissue adhesion, and high background noise. To address these, this study introduces a lung tumor segmentation method based on Transformer and attention mechanisms. In the Transformer encoder stage, both global and local attention mechanisms are incorporated to enable the network to simultaneously focus on both global and local contextual information. In the skip connection stage, a channel-prior convolutional attention mechanism is utilized to enhance the spatial perception ability for complex lesions and reduce the channel dimension redundancy, such that the tumor segmentation accuracy can be improved. The experimental results on the private GDPH and public LUNG1 datasets demonstrate that the proposed method outperforms eight comparative methods in terms of the Dice metric by achieving approximately 90.96% and 88.18% on the two datasets, respectively. The proposed method can provide reliable assistance for clinical diagnosis and treatment.

Key words: lung tumor, medical image segmentation, convolutional neural networks, Transformer, attention mechanisms

中图分类号: 

  • TP391.4
[1] SIEGEL R L, MILLER K D, WAGLE N S, et al. Cancer statistics, 2023[J]. CA: A Cancer Journal for Clinicians, 2023, 73(1): 17-48.
[2] KASINATHAN G, JAYAKUMAR S, GANDOMI A H. Automated 3-D lung tumor detection and classification by an active contour model and CNN classifier[J]. Expert Systems with Applications, 2019, 134: 112-119.
[3] 刘桂红, 赵亮, 孙劲光, 等. 一种改进粒子群优化算法的Otsu图像阈值分割方法[J]. 计算机科学, 2016, 43(3): 309-312.
LIU G H, ZHAO L, SUN J G, et al. Otsu image threshold segmentation method based on improved particle swarm optimization[J]. Computer Science, 2016, 43(3): 309-312.
[4] 贾鹤鸣, 李瑶, 姜子超, 等. 基于改进共生生物搜索算法的林火图像多阈值分割[J]. 计算机应用, 2021, 41(5): 1465-1470.
JIA H M, LI Y, JIANG Z C, et al. Multi-threshold segmentation of forest fire images based on modified symbiotic organisms search algorithm[J]. Journal of Computer Applications, 2021, 41(5): 1465-1470.
[5] MALATHI M, SINTHIA P, JALALDEEN K. Active contour based segmentation and classification for pleura diseases based on Otsu’s thresholding and support vector machine (SVM) [J]. Asian Pacific Journal of Cancer Prevention: APJCP, 2019, 20(1): 167.
[6] GOPI K, SELVAKUMAR J. Lung tumor area recognition and classification using EK-mean clustering and SVM[C]//2017 International Conference on Nextgen Electronic Technologies: Silicon to Software (ICNETS2) . Tamil Nadu: IEEE, 2017: 97-100.
[7] SANGAMITHRAA P B, GOVINDARAJU S. Lung tumour detection and classification using EK-Mean clustering[C]//2016 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET) . Chennai : IEEE, 2016: 2201-2206.
[8] MI H, PETITJEAN C, DUBRAY B, et al. Automatic lung tumor segmentation on PET images based on random walks and tumor growth model[C]//2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI) . Beijing: IEEE, 2014: 1385-1388.
[9] JONES J L, XIE X, ESSA E. Combining region-based and imprecise boundary-based cues for interactive medical image segmentation[J]. International Journal for Numerical Methods in Biomedical Engineering, 2014, 30(12): 1649-1666.
[10] YANG B, XIANG D, YU F, et al. Lung tumor segmentation based on the multi-scale template matching and region growing[C]//Medical Imaging 2018: Biomedical Applications in Molecular, Structural, and Functional Imaging. Houston: SPIE, 2018, 10578: 645-651.
[11] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Santiago: IEEE, 2015: 3431-3440.
[12] RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Munich: Springer International Publishing, 2015: 234-241.
[13] DUTANDE P, BAID U, TALBAR S. Deep residual separable convolutional neural network for lung tumor segmentation[J]. Computers in Biology and Medicine, 2022, 141: 105161.
[14] YANG J, WU B, LI L, et al. MSDS-UNet: a multi-scale deeply supervised 3D U-Net for automatic segmentation of lung tumor in CT[J]. Computerized Medical Imaging and Graphics, 2021, 92: 101957.
[15] ZHANG G, YANG Z, JIANG S. Automatic lung tumor segmentation from CT images using improved 3D densely connected UNet[J]. Medical & Biological Engineering & Computing, 2022, 60(11): 3311-3323.
[16] KAMAL U, RAFI A M, HOQUE R, et al. Lung cancer tumor region segmentation using recurrent 3d-denseunet[C]//Thoracic Image Analysis: Second International Workshop, TIA 2020. Lima: Springer International Publishing, 2020: 36-47.
[17] DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16×16 words: Transformers for image recognition at scale[EB/OL]. arXiv：2010. 11929 (2021-06-03) [2023-11-11]. https://doi.org/10.48550.
[18] CHENG B, MISRA I, SCHWING A G, et al. Masked-attention mask transformer for universal image segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 1290-1299.
[19] HATAMIZADEH A, TANG Y, NATH V, et al. Unetr: transformers for 3D medical image segmentation[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. New Orleans : IEEE, 2022: 574-584.
[20] WANG S, JIANG A, LI X, et al. DPBET: a dual-path lung nodules segmentation model based on boundary enhancement and hybrid transformer[J]. Computers in Biology and Medicine, 2022, 151: 106330.
[21] WANG X, LI Y. STC-Net: fusing swin transformer and convolution neural network for 2D medical Image segmentation[C]//2022 2nd International Conference on Electronic Information Engineering and Computer Technology (EIECT) . Xian: IEEE, 2022: 317-320.
[22] WANG L, LI R, ZHANG C, et al. UNetFormer: a unet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, 190: 196-214.
[23] CHEN J, LU Y, YU Q, et al. TransUNet: transformers make strong encoders for medical image segmentation[EB/OL]. arXiv: 2102.04306(2021-02-08) [2023-11-11]. https://arxiv.org/abs/2102.04306.
[24] BJORCK J, GOMES C, SELMAN B, et al. Understanding batch normalization[C]// NIPS’18: Proceedings of the 32nd International Conference on Neural Information Processing Systems. Red Hook, New York: Curran Associates Incorporation, 2018: 7705–7716.
[25] CAI Y, WANG Y. Ma-unet: an improved version of unet based on multi-scale and attention mechanism for medical image segmentation[C]//Third International Conference on Electronics and Communication; Network and Computer Technology (ECNCT 2021) . Xiamen: SPIE, 2022, 12167: 205-211.
[26] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Vancouver : IEEE, 2018: 7132-7141.
[27] WOO S, PARK J, LEE J Y, et al. Cbam: convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV) . Munich: Springer, 2018: 3-19.
[28] ZHU X, CHENG D, ZHANG Z, et al. An empirical study of spatial attention mechanisms in deep networks[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. California: IEEE, 2019: 6688-6697.
[29] HUANG H, CHEN Z, ZOU Y, et al. Channel prior convolutional attention for medical image segmentation[EB/OL]. arXiv: 2306.05196 (2023-06-08) [2023-11-11]. https://arxiv.org/abs/2306.05196.
[30] AERTS H J W L, VELAZQUEZ E R, LEIJENAAR R T H, et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach[J]. Nature Communications, 2014, 5(1): 4006.
[31] CAO H, WANG Y, CHEN J, et al. Swin-unet: unet-like pure transformer for medical image segmentation[C]//European Conference on Computer Vision. Cham: Springer Nature Switzerland, 2022: 205-218.
[32] 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.
[33] VALANARASU J M J, SINDAGI V A, HACIHALILOGLU I, et al. Kiu-net: overcomplete convolutional architectures for biomedical image and volumetric segmentation[J]. IEEE Transactions on Medical Imaging, 2021, 41(4): 965-976.
[34] LI C, QIANG Y, SULTAN R I, et al. FocalUNETR: a focal transformer for boundary-aware prostate segmentation using CT images[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer Nature Switzerland, 2023: 592-602.
[35] CHEN W, WEI H, PENG S, et al. HSN: hybrid segmentation network for small cell lung cancer segmentation[J]. IEEE Access, 2019, 7: 75591-75603.
[36] LE V L, SAUT O. Rrc-unet 3D for lung tumor segmentation from CT scans of non-small cell lung cancer patients[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops. Paris: IEEE, 2023: 2316-2325.
[1] 谢伟立, 张军. 一种基于多尺度的多层卷积稀疏编码网络[J]. 广东工业大学学报, 2024, 41(06): 125-132.
[2] 欧嘉俊, 曾伟良, 李谕锋, 范竞敏. 基于多智能体注意力机制的自动巡检路线强化学习模型[J]. 广东工业大学学报, 2024, 41(05): 39-47,71.
[3] 冯广, 鲍龙. 基于红外可见光融合的复杂环境下人脸识别方法[J]. 广东工业大学学报, 2024, 41(03): 62-70,109.
[4] 李雪森, 谭北海, 余荣, 薛先斌. 基于YOLOv5的轻量化无人机航拍小目标检测算法[J]. 广东工业大学学报, 2024, 41(03): 71-80.
[5] 涂泽良, 程良伦, 黄国恒. 基于局部正交特征融合的小样本图像分类[J]. 广东工业大学学报, 2024, 41(02): 73-83.
[6] 郭傲, 许柏炎, 蔡瑞初, 郝志峰. 基于时序对齐的风格控制语音合成算法[J]. 广东工业大学学报, 2024, 41(02): 84-92.
[7] 杨镇雄, 谭台哲. 基于生成对抗网络的低光照图像增强算法[J]. 广东工业大学学报, 2024, 41(01): 55-62.
[8] 赖志茂, 章云, 李东. 基于Transformer的人脸深度伪造检测技术综述[J]. 广东工业大学学报, 2023, 40(06): 155-167.
[9] 曾安, 陈旭宙, 姬玉柱, 潘丹, 徐小维. 基于自注意力和三维卷积的心脏多类分割方法[J]. 广东工业大学学报, 2023, 40(06): 168-175.
[10] 吴亚迪, 陈平华. 基于用户长短期偏好和音乐情感注意力的音乐推荐模型[J]. 广东工业大学学报, 2023, 40(04): 37-44.
[11] 黄晓湧, 李伟彤. 基于TSSI和STB-CNN的跌倒检测算法[J]. 广东工业大学学报, 2023, 40(04): 53-59.
[12] 曹智雄, 吴晓鸰, 骆晓伟, 凌捷. 融合迁移学习与YOLOv5的安全帽佩戴检测算法[J]. 广东工业大学学报, 2023, 40(04): 67-76.
[13] 赖东升, 冯开平, 罗立宏. 基于多特征融合的表情识别算法[J]. 广东工业大学学报, 2023, 40(03): 10-16.
[14] 谢国波, 林立, 林志毅, 贺笛轩, 文刚. 基于YOLOv4-MP的绝缘子爆裂缺陷检测方法[J]. 广东工业大学学报, 2023, 40(02): 15-21.
[15] 吴俊贤, 何元烈. 基于通道注意力的自监督深度估计方法[J]. 广东工业大学学报, 2023, 40(02): 22-29.
Viewed
Full text
317
HTML PDF
Just accepted Online first Issue Just accepted Online first Issue
0 0 0 0 11 306

  From Others local
  Times 49 268
  Rate 15% 85%

Abstract
172
Just accepted Online first Issue
0 17 155
  From local
  Times 172
  Rate 100%

Cited
Web of Science  Crossref   ScienceDirect  Search for Citations in Google Scholar >>
 
This page requires you have already subscribed to WoS.
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发