An Auxiliary Information Embedding Network for Tuberculosis Classification in X-Ray Images

AN Le; YANG Jing; PENG Ke-xin

doi:10.12263/newIT.2024.05.004

您当前的位置：

首页 >

文章列表页 >

An Auxiliary Information Embedding Network for Tuberculosis Classification in X-Ray Images

RESEARCH PAPER | 更新时间：2025-05-23

- An Auxiliary Information Embedding Network for Tuberculosis Classification in X-Ray Images
- New Generation of Information Technology Vol. 7, Issue 5, Pages: 16-25(2024)
- 作者机构：
  
  1.南充文化旅游职业学院，四川南充 637400
  2.成都理工大学，四川成都 610059
- 作者简介：
- 基金信息：
- DOI：10.12263/newIT.2024.05.004
  CLC： TP391.41
- Accepted：12 March 2025，
  
  Published：15 May 2024
- 稿件说明：
移动端阅览
安乐, 杨静, 彭柯鑫. 一种嵌入辅助信息的轻量化肺结核胸片分类算法[J]. 新一代信息技术, 2024, 7(5): 16-25

AN Le, YANG Jing, PENG Ke-xin. An Auxiliary Information Embedding Network for Tuberculosis Classification in X-Ray Images[J]. New Generation of Information Technology, 2024, 7(5): 16-25
安乐, 杨静, 彭柯鑫. 一种嵌入辅助信息的轻量化肺结核胸片分类算法[J]. 新一代信息技术, 2024, 7(5): 16-25 DOI： 10.12263/newIT.2024.05.004.

AN Le, YANG Jing, PENG Ke-xin. An Auxiliary Information Embedding Network for Tuberculosis Classification in X-Ray Images[J]. New Generation of Information Technology, 2024, 7(5): 16-25 DOI： 10.12263/newIT.2024.05.004.

摘要

卷积神经网络在分类肺结核患者的X光胸片中具有突出表现，但在平衡网络复杂度和识别精度上依然具有挑战性。针对上述问题，本文提出了一种嵌入辅助信息的轻量级胸片分类算法（Tuberculosis Auxiliary Information Embedding Network，TAIE-Net），以实现低复杂度情况下的准确分类。首先，使用ResNet残差网络作为基本结构，对其轻量化设计后作为网络特征提取主干。然后，对X光胸片内的左、右肺关系进行建模得到区域信息，再对各胸片之间的肺组织进行建模得到差异信息。最后，将区域信息和差异信息整合，形成补偿信息并嵌入轻量化分类网络的优化过程中，使深层语义信息更具鉴别力，提升了模型对关键特征的筛选能力。在肺结核公开数据集上进行分类实验，所提方法的分类准确率为93.8%，敏感度为92.5%，

分数为93.7%。该方法相较于Mobile-Former和Mobile-ViT等轻量化网络，有效提升了肺结核胸片分类的精度，且在网络推理效率、参数量、权重方面更具有优势。

Abstract

The convolutional neural network has a prominent performance in classifying pulmonary tuberculosis patients in X-ray images

but it is still challenging to balance the complexity and recognition accuracy. To solve the above problems

a lightweight chest film classification network TAIE-Net embedded with auxiliary information is proposed to achieve accurate classification under low complexity. Firstly

the ResNet is used as the basic structure

and after lightweight design

it serves as the backbone for network feature extraction. Then

model the relationship between the left and right lungs in the X-ray chest radiography to obtain regional information

and then model the lung tissue between each chest film to obtain differential information. Finally

the regional and difference information are integrated to form auxiliary information and embedded in the classification network to make the deep semantic information more discriminative. The classification experiment is conducted on the public dataset of pulmonary tuberculosis

and the proposed method classification acc

uracy reaches 93.8%

the sensitivity reached 92.5%

and the

score reaches 93.7%. This method effectively improves the accuracy of pulmonary tuberculosis chest X-ray classification compared to lightweight networks such as MobileFormer and MobileViT and has more advantages in model computing complexity

parameter quantity

and weight size.

关键词

Keywords

references

World Health Organization . Global tuberculosis report 2021 [EB/OL ] . ( 2021-10-14 ) [ 2023-12-15 ] . https://www.who.int/publications/i/item/9789240037021 https://www.who.int/publications/i/item/9789240037021 .

安乐 , 彭柯鑫 , 杨兴 , 等 . 一种融合肺炎影像特征的肺结核病灶检测迁移网络 [J ] . 光学技术 , 2023 , 49 ( 1 ): 120 - 128 .

AN L , PENG K X , YANG X , et al . A transfer network for pulmonary tuberculosis lesions detection on the fusion of image features of pneumonia [J ] . Optical Technique , 2023 , 49 ( 1 ): 120 - 128 . (in Chinese)

胡恒豪 , 王俊峰 , 方智阳 , 等 . 基于深层迁移学习的DR胸片肺结核病灶检测 [J ] . 四川大学学报: 自然科学版 , 2020 , 57 ( 3 ): 459 - 468 .

HU H H , WANG J F , FANG Z Y , et al . Deep transfer learning based pulmonary tuberculosis lesions detection on DR films [J ] . Journal of Sichuan University: Natural Science Edition , 2020 , 57 ( 3 ): 459 - 468 . (in Chinese)

AN L , PENG K , YANG X , et al . E-TBNet: light deep neural network for automatic detection of tuberculosis with X-ray DR imaging [J ] . Sensors: Basel , 2022 , 22 ( 3 ): 821 .

王旭 , 段辉宏 , 聂生东 . 基于CT影像组学的非小细胞肺癌预后分析方法 [J ] . 电子学报 , 2020 , 48 ( 4 ): 637 - 642 .

WANG X , DUAN H H , NIE S D . Prognostic analysis method for non-small cell lung cancer based on CT radiomics [J ] . Acta Electronica Sinica , 2020 , 48 ( 4 ): 637 - 642 . (in Chinese)

XIAO A Q , SHEN B L , SHI X J , et al . Intraoperative glioma grading using neural architecture search and multi-modal imaging [J ] . IEEE Transactions on Medical Imaging , 2022 , 41 ( 10 ): 2570 - 2581 .

HUANG B S , TIAN J R , ZHANG H Y , et al . Deep semantic segmentation feature-based radiomics for the classification tasks in medical image analysis [J ] . IEEE Journal of Biomedical and Health Informatics , 2021 , 25 ( 7 ): 2655 - 2664 .

CHEN X , KUANG T S , DENG H , et al . Dual adversarial attention mechanism for unsupervised domain adaptive medical image segmentation [J ] . IEEE Transactions on Medical Imaging , 2022 , 41 ( 11 ): 3445 - 3453 .

张淑军 , 彭中 , 李辉 . SAU-Net: 基于U-Net和自注意力机制的医学图像分割方法 [J ] . 电子学报 , 2022 , 50 ( 10 ): 2433 - 2442 .

ZHANG S J , PENG Z , LI H . SAU-Net: Medical image segmentation method based on U-Net and self-attention [J ] . Acta Electronica Sinica , 2022 , 50 ( 10 ): 2433 - 2442 . (in Chinese)

赖小波 , 许茂盛 , 徐小媚 . 基于多模型融合和区域迭代生长的视网膜血管自动分割 [J ] . 电子学报 , 2019 , 47 ( 12 ): 2611 - 2621 .

LAI X B , XU M S , XU X M . Automatic retinal vessel segmentation based on multi-model fusion and region iterative growth [J ] . Acta Electronica Sinica , 2019 , 47 ( 12 ): 2611 - 2621 . (in Chinese)

MA Y H , LIU J , LIU Y H , et al . Structure and illumination constrained GAN for medical image enhancement [J ] . IEEE Transactions on Medical Imaging , 2021 , 40 ( 12 ): 3955 - 3967 .

LEVIN D R , STELLER J , ANDERSON A , et al . Enabling human space exploration missions through progressively earth independent medical operations (EIMO) [J ] . IEEE Open Journal of Engineering in Medicine and Biology , 2023 , 4 : 162 - 167 .

MOZAFFARZADEH M , MINONZIO C , DE JONG N , et al . Lamb waves and adaptive beamforming for aberration correction in medical ultrasound imaging [J ] . IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control , 2021 , 68 ( 1 ): 84 - 91 .

RAJARAMAN S , ZAMZMI G , FOLIO L , et al . Chest X-ray Bone Suppression for Improving Classification of Tuberculosis-Consistent Findings [J ] . Diagnostics: Basel , 2021 , 11 ( 5 ): 840 .

RAHMAN T , KHANDAKAR A , KADIR M A , et al . Reliable tuberculosis detection using chest X-ray with deep learning, segmentation and visualization [J ] . IEEE Access , 2020 , 8 : 191586 - 191601 .

LAKHANI P , SUNDARAM B . Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks [J ] . Radiology , 2017 , 284 ( 2 ): 574 - 582 .

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 (CVPR) . IEEE , 2016 : 770 - 778 .

WANG Q L , WU B G , ZHU P F , et al . ECA-net: efficient channel attention for deep convolutional neural networks [C ] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . IEEE , 2020 : 11531 - 11539 .

LI Z , WANG C , HAN M , et al . Thoracic disease identification and localization with limited supervision [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 8290 - 8299 .

JAEGER S , CANDEMIR S , ANTANI S , et al . Two public chest X-ray datasets for computer-aided screening of pulmonary diseases [J ] . Quant Imaging Med Surg , 2014 , 4 ( 6 ): 475 - 477 .

MEHTA S , RASTEGARI M . MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer [EB/OL ] . ( 2021 )[2024 ] . https://arxiv.org/abs/2110.02178v2 https://arxiv.org/abs/2110.02178v2 .

CHEN Y P , DAI X Y , CHEN D D , et al . Mobile-former: Bridging MobileNet and transformer [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 5260 - 5269 .

PAN J T , BULAT A , TAN F W , et al . EdgeViTs: Competing light-weight CNNs on mobile devices with vision transformers [M ] // Computer Vision-ECCV 2022 . Cham : Springer Nature Switzerland , 2022 : 294 - 311 .

WANG W H , XIE E Z , LI X , et al . PVT v2: Improved baselines with pyramid vision transformer [J ] . Computational Visual Media , 2022 , 8 ( 3 ): 415 - 424 .

SANDLER M , HOWARD A , ZHU M L , et al . MobileNetV2: Inverted residuals and linear bottlenecks [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 4510 - 4520 .

IANDOLA F N , HAN S , MOSKEWICZ M W , et al . Squee-zeNet: Alexnet-level accuracy with 50x fewer parameters and <0.5MB model size [EB/OL ] . ( 2016 )[2024 ] . https://arxiv.org/abs/1602.07360v4 https://arxiv.org/abs/1602.07360v4 .

MA N N , ZHANG X Y , ZHENG H T , et al . ShuffleNet V2: Practical guidelines for efficient CNN architecture design [M ] // Computer Vision - ECCV 2018 . Cham : Springer International Publishing , 2018 : 122 - 138 .

MEHTA S , RASTEGARI M , SHAPIRO L , et al . ESPNetv2: A light-weight, power efficient, and general purpose convolutional neural network [C ] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2019 : 9182 - 9192 .

ASTA D M . Kernel density estimation on symmetric spaces of non-compact type [J ] . Journal of Multivariate Analysis , 2021 , 181 : 104676 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

An Efficient Design Method Based on Artificial Neural Network for EIT Metasurfaces

Related Author

YUAN Lin

TANG Jia

YU Hai-yang

ZHANG Cheng

Yu Haiyang

Related Institution

Southwest China Research Institute of Electronic Equipment

⁰