计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 230900116-6.doi: 10.11896/jsjkx.230900116

• 图像处理&多媒体技术 • 上一篇    下一篇

基于注意力机制和双分支网络的胸部疾病分类

宋子岩1, 罗川1, 李天瑞2, 陈红梅2   

  1. 1 四川大学计算机学院 成都 610065
    2 西南交通大学计算机与人工智能学院 成都 611756
  • 出版日期:2024-11-16 发布日期:2024-11-13
  • 通讯作者: 罗川(cluo@scu.edu.cn)
  • 作者简介:(zy_song2150@163.com)
  • 基金资助:
    国家自然科学基金(62076171,62376230),四川省自然科学基金(2022NSFSC0898)

Classification of Thoracic Diseases Based on Attention Mechanisms and Two-branch Networks

SONG Ziyan1, LUO Chuan1, LI Tianrui2, CHEN Hongmei2   

  1. 1 College of Computer Science,Sichuan University,Chengdu 610065,China
    2 School of Computing and Artificial Intelligence,Southwest Jiaotong University,Chengdu 611756,China
  • Online:2024-11-16 Published:2024-11-13
  • About author:SONG Ziyan,born in 1999,postgra-duate.His main research interest is medical image processing and analysis.
    LUO Chuan,born in 1987,Ph.D,asso-ciate professor,Ph.D supervisor,is a member of CCF(No.45031M).His mainresearch interests include data mining and machine learning.
  • Supported by:
    National Natural Science Foundation of China(62076171,62376230) and Natural Science Foundation of Sichuan Province,China(2022NSFSC0898).

PDF (PC)

166

摘要： 基于胸部X光片的胸部疾病分类对于提高诊断准确率、减轻医疗卫生系统压力具有重要意义。不同胸部疾病的发病区域尺寸存在巨大差异,是基于胸部X光片的胸部疾病分类任务面临的主要挑战。在对发病区域较小的疾病进行分类时,图像中大部分区域均为噪声区域,传统方法难以有效应对疾病间巨大的尺寸差异。针对这一问题,提出了一种结合多尺度特征的掩码构造方法,以DenseNet-121为特征提取器,构建了一个双分支网络,使用全局网络进行总体分类,并将微小病变区域送入局部分支以减轻噪声区域的干扰,最终利用基于注意力机制分支特征融合模块自适应地融合两个分支的分类特征信息。在ChestX-ray14数据集上进行了对比实验、消融实验和参数敏感性分析,结果表明,所提方法对14种胸部疾病分类的平均AUC高于现有方法,具有有效性且对参数不敏感。

关键词: 胸部疾病分类, 深度学习, 胸部X光片, 注意力机制, 特征融合

Abstract: Thoracic disease classification based on chest radiographs is important to improve diagnostic accuracy and reduce the pressure on the healthcare system.The huge variation in the size of the regions of different thoracic diseases is the main challenge in the chest radiograph-based classification of thoracic diseases.When classifying diseases with small onset regions,most of the regions in the image are noisy regions,and it is difficult for traditional methods to cope with the huge size differences among di-seases effectively.To address this problem,a mask construction method combining multi-scale features is proposed,using DenseNet-121 as the feature extractor,a two-branch network is constructed,in which the global network is used for the overall classification,and tiny lesion regions are fed into the local branches to mitigate the interference of noisy regions.The branch feature fusion module based on the attentional mechanism is used to fuse the classification features from the two branches' information adaptively.Comparison experiments,ablation experiments,and parameter sensitivity analyses are performed on the ChestX-ray14 dataset.The experimental results show that the average AUC of the proposed method for classifying 14 thoracic diseases is higher than that of the existing methods,which is effective and parameter insensitive.

Key words: Thoracic disease classification, Deep learning, Chest radiographs, Attention mechanism, Feature fusion

中图分类号: 

  • TP391
[1]MA X Q,ZHU L L,KURCHE J S,et al.Global and regionalburden of interstitial lung disease and pulmonary sarcoidosis from 1990 to 2019:results from the global burden of disease study 2019[J].Thorax,2022,77(6):596-605.
[2]MCLAUGHLIN L,BOND R,HUGHES C,et al.Computing eyegaze metrics for the automatic assessment of radiographer performance during x-ray image interpretation [J].International Journal of Medical Informatics,2017,105:11-21.
[3]HOSSAIN M I,HOSSAIN S M,ZUNAED M,et al.A novel attention mechanism using anatomical prior probability maps for thoracic disease classification from x-ray images [J].arXiv:2210.02998,2022.
[4]WANG H Y,WANG S S,QIN Z B,et al.Triple attention lear-ning for classification of 14 thoracic diseases using chest radiography[J].Medical Image Analysis,2021,67:101846.
[5]GUAN Q J,HUANG Y P,ZHONG Z,et al.Thorax diseaseclassification with attention guided convolutional neural network[J].Pattern Recognition Letters,2020,131:38-45.
[6]ÇALLI E,SOGANCIOGLU E,VAN G B,et al.Deep learning for chest x-ray analysis:a survey [J].Medical Image Analysis,2021,72:102125.
[7]WANG X S,PENG Y F,LU L,et al.Chestx-ray8:hospital-scale chest x-ray database andbenchmarks on weakly-supervised classification and localization of common thorax diseases[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:2097-2106.
[8]RAJPURKAR P,IRVINJ,ZHU K,et al.Chexnet:radiologist-level pneumonia detection on chest x-rays with deep learning [J].arXiv:1711.05225,2017.
[9]HUANG G,LIU Z,VAN DER MAATEN L,et al.Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:4700-4708.
[10]IOFFEG S,SZEGEDY C.Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//International Conference on Machine Learning.2015:448-456.
[11]BALTRUSCHAT I M,NICKISCH H,GRASS M,et al.Com-parison of deep learning approaches for multi-label chest x-ray classification[J].Scientific Reports,2019,9(1):6381.
[12]GUAN Q J,HUANG Y P.Multi-label chest x-ray image classification via category-wise residual attention learning [J].Pattern Recognition Letters,2020,130:259-266.
[13]GE Z Y,MAHAPATRA D,CHANG X J,et al.Improvingmulti-label chest x-ray disease diagnosis by exploiting disease and health labels dependencies [J].Multimedia Tools and Applications,2020,79:14889-14902.
[14]CAI J Z,LU L,HARRISON A P,et al.Iterative attention mi-ning for weakly supervised thoracic disease pattern localization in chest x-rays[C]// Medical Image Computing and Computer Assisted Intervention.2018:589-598.
[15]LI B C,KANG G X,CHENG K,et al.Attention-guided convolutional neural network for detecting pneumonia on chest x-rays[C]//2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society.2019:4851-4854.
[16]HU J,SHEN L,SUN G.Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on ComputerVision and Pattern Recognition.2018:7132-7141.
[17]SAEDNIA K,JALALIFAR A,EBRAHIMI S,et al.An atten-tion-guided deep neural network for annotating abnormalities in chestx-ray images:visualization of network decision basis[C]//2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society.2020:1258-1261.
[18]WANG H Y,JIA H Z,LU L,et al.Thorax-net:an attentionregularized deep neural network for classification of thoracic diseases on chest radiography [J].IEEE Journal of Biomedical and Health Informatics,2019,24(2):475-485.
[19]KAMAL U,ZUNAED M,NIZAM N B,et al.Anatomy-xnet:an anatomy aware convolutional neural network for thoracic disease classification in chest x-rays [J].IEEE Journal of Biomedical and Health Informatics,2022,26(11):5518-5528.
[20]CHEN B Z,ZHANG Z,LIN J Y,et al.Two-stream collaborativenetwork for multi-label chest x-ray Image classification with lung segmentation [J].Pattern Recognition Letters,2020,135:221-227.
[21]RUSSAKOVSKY O,DENG J,SU H,et al.Imagenet large scale visual recognition challenge [J].International Journal of Computer Vision,2015,115:211-252.
[22]WOO S,PARK J,LEE J Y,et al.Cbam:convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision.2018:3-19.
[23]YAO L,POBLENZE,DAGUNTS D,et al.Learning to diagnose from scratch by exploiting dependencies among labels [J].ar-Xiv:1710.10501,2017.
[24]LI Z,WANG C,HAN M,et al.Thoracic disease identification and localization with limited supervision[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:8290-8299.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发