DRT Net：面向特征增强的双残差Res-Transformer肺炎识别模型

周涛; 彭彩月; 杜玉虎; 党培; 刘凤珍; 陆惠玲

您当前的位置：

首页 >

文章列表页 >

DRT Net：面向特征增强的双残差Res-Transformer肺炎识别模型

更新时间：2024-01-23

- DRT Net：面向特征增强的双残差Res-Transformer肺炎识别模型
- DRT Net： dual res-transformer pneumonia recognition model oriented to feature enhancement
- 光学精密工程 2024年页码：1-13
- 作者机构：
  
  1.北方民族大学计算机科学与工程学院，宁夏银川 750021
  2.北方民族大学图像图形智能处理国家民委重点实验室，宁夏银川 750021
  3.宁夏医科大学医学信息与工程学院，宁夏银川 750004
- 作者简介：
  
  [ "周涛（1977-），男，宁夏同心人。博士，教授，2010年于西北工业大学获得博士学位，主要从事医学图像分析处理、深度学习、模式识别等方面的研究。E-mail： zhoutaonxmu@126.com" ]
- 基金信息：
  
  国家自然科学基金(62062003);宁夏自然科学基金资助项目(2022AAC03149)
- DOI：
  中图分类号：
扫描看全文
周涛,彭彩月,杜玉虎等.DRT Net：面向特征增强的双残差Res-Transformer肺炎识别模型[J].光学精密工程,

ZHOU Tao,PENG Caiyue,DU Yuhu,et al.DRT Net： dual res-transformer pneumonia recognition model oriented to feature enhancement[J].Optics and Precision Engineering,
周涛,彭彩月,杜玉虎等.DRT Net：面向特征增强的双残差Res-Transformer肺炎识别模型[J].光学精密工程, DOI：10.37188/OPE.XXXXXXXX.0001

ZHOU Tao,PENG Caiyue,DU Yuhu,et al.DRT Net： dual res-transformer pneumonia recognition model oriented to feature enhancement[J].Optics and Precision Engineering, DOI：10.37188/OPE.XXXXXXXX.0001

摘要

目的基于深度学习的肺部X射线图像识别是近些年的研究热点。但是由于肺部X射线图像的病灶区域较小、形状复杂，与正常组织间的边界模糊，使得肺炎图像中的病灶特征提取不充分。方法本文提出了一个面向特征增强的双残差Res-Transformer肺炎识别模型，设计三种不同的特征增强策略对模型特征提取能力进行增强。该模型的主要工作有以下几个方面：首先，设计了组注意力双残差模块（GADRM），采用双残差结构进行高效的特征融合，将双残差结构与通道混洗、通道注意力、空间注意力结合，增强模型对于病灶区域特征的提取能力；其次，在网络的高层采用全局局部特征提取模块（GLFEM），结合CNN和Transformer的优势使网络充分提取图像的全局和局部特征，获得高层语义信息的全局特征，进一步增强网络的语义特征提取能力；然后，设计了跨层双注意力特征融合模块（CDAFFM），融合浅层网络的空间信息以及深层网络的通道信息，对网络提取到的跨层特征进行增强。结果为了验证本文模型的有效性，分别在COVID-19 CHEST X-RAY数据集上进行消融实验和对比实验，实验结果表明，本文所提出网络的准确率、精确率、召回率，F1值和AUC值分别为98.41%，94.42%，94.20%，94.26%和99.65%。结论本文所提出的DRT Net能够帮助放射科医生使用胸部X光片对肺炎进行诊断，具有重要的临床作用。

Abstract

ObjectiveDeep learning based on lung X-ray images recognition is a hot research topic recently. Due to the small size and complex shape of the lesion area in the lung X-ray images， the boundary between the lesion area and the normal tissue is blurred， which makes the feature extraction of lesion inadequate in pneumonia images.MethodDual Res-Transformer pneumonia recognition model oriented to feature enhancement is proposed in this paper. Three different feature enhancement strategies are designed to enhance the feature extraction ability of the model. The main works of the model are as follows： Firstly， the Group Attention Dual Residual Module （GADRM） is designed to employ the dual-residual structure for efficient feature fusion and combine the dual-residual structure with channel shuffle， channel attention and spatial attention to enhance local region feature extraction capability. Secondly， the Global-Local Feature Extraction Module （GLFEM） is employed at the higher level of the network， which combines the advantages of CNN and Transformer to fully extract the global and local feature of images and obtains global feature with high-level semantic information， and further enhances the semantic feature extraction ability of the network. Thirdly， the Cross-layer Dual Attention Feature Fusion Module （CDAFFM） is designed to fuse the spatial information of shallow network and channel information of the deep network， which can enhance the cross-layer features extracted by the network.ResultAblation experiments and comparison experiments are conducted on the COVID-19 CHEST X-RAY dataset respectively to verify the effectiveness of the model in this paper. The experimental results show that the accuracy rate， precision rate， recall rate， F1 value and AUC value of the proposed network in this paper are 98.41%， 94.42%， 94.20%， 94.26%， and 99.65%. respectively.ConclusionThe model can help radiologists to diagnose different pneumonia cases using chest X-rays and play an important clinical role in pneumonia computer aided diagnosis.

关键词

肺炎识别X射线图像特征增强双残差结构Transformer

Keywords

Pneumonia recognitionX-ray imageFeature enhancementDual residual modelTransformer

references

ZHOU T， LIU F Z， LU H L， et al. A review of deep learning imaging diagnostic methods for COVID-19［J］. Electronics， 2023， 12（5）： 1167. doi: 10.3390/electronics12051167http://dx.doi.org/10.3390/electronics12051167

TORRES-RAMIREZ C A， TIMARAN-MONTENEGRO D， MATEO-CAMACHO Y S， et al. CT-based pathological lung opacities volume as a predictor of critical illness and inflammatory response severity in patients with COVID-19［J］. Heliyon， 2022， 8（12）： e11908. doi: 10.1016/j.heliyon.2022.e11908http://dx.doi.org/10.1016/j.heliyon.2022.e11908

REN K， GU Y， LUO M S， et al. Deep-learning-based denoising of X-ray differential phase and dark-field images［J］. European Journal of Radiology， 2023， 163： 110835. doi: 10.1016/j.ejrad.2023.110835http://dx.doi.org/10.1016/j.ejrad.2023.110835

MAMALAKIS M， SWIFT A J， VORSELAARS B， et al. DenResCov-19： a deep transfer learning network for robust automatic classification of COVID-19， pneumonia， and tuberculosis from X-rays［J］. Computerized Medical Imaging and Graphics， 2021， 94： 102008. doi: 10.1016/j.compmedimag.2021.102008http://dx.doi.org/10.1016/j.compmedimag.2021.102008

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）. June 27-30， 2016. Las Vegas， NV， USA. IEEE， 2016： 770-778. doi: 10.1109/cvpr.2016.90http://dx.doi.org/10.1109/cvpr.2016.90

ZHOU T， CHANG X Y， LIU Y C， et al. COVID-ResNet： COVID-19 recognition based on improved attention ResNet［J］. Electronics， 2023， 12（6）： 1413. doi: 10.3390/electronics12061413http://dx.doi.org/10.3390/electronics12061413

CHEN Y F， LIN Y L， XU X D， et al. Classification of lungs infected COVID-19 images based on inception-ResNet［J］. Computer Methods and Programs in Biomedicine， 2022， 225： 107053. doi: 10.1016/j.cmpb.2022.107053http://dx.doi.org/10.1016/j.cmpb.2022.107053

HUANG Q H， LEI Y， XING W Y， et al. Evaluation of pulmonary edema using ultrasound imaging in patients with COVID-19 pneumonia based on a non-local channel attention ResNet［J］. Ultrasound in Medicine & Biology， 2022， 48（5）： 945-953. doi: 10.1016/j.ultrasmedbio.2022.01.023http://dx.doi.org/10.1016/j.ultrasmedbio.2022.01.023

AHILA T， SUBHAJINI A C. E-GCS： detection of COVID-19 through classification by attention bottleneck residual network［J］. Engineering Applications of Artificial Intelligence， 2022， 116： 105398. doi: 10.1016/j.engappai.2022.105398http://dx.doi.org/10.1016/j.engappai.2022.105398

CHEN B Z， LI J X， GUO X B， et al. DualCheXNet： dual asymmetric feature learning for thoracic disease classification in chest X-rays［J］. Biomedical Signal Processing and Control， 2019， 53： 101554. doi: 10.1016/j.bspc.2019.04.031http://dx.doi.org/10.1016/j.bspc.2019.04.031

吴宣言，缑新科，朱子重，等. 深层聚合残差密集网络的超声图像左心室分割［J］. 中国图象图形学报， 2020， 25（9）： 1930-1942. doi: 10.11834/jig.190552http://dx.doi.org/10.11834/jig.190552

WU X Y， GOU X K， ZHU Z Z， et al. Left ventricular segmentation on ultrasound images using deep layer aggregation for residual dense networks［J］. Journal of Image and Graphics， 2020， 25（9）： 1930-1942.（in Chinese）. doi: 10.11834/jig.190552http://dx.doi.org/10.11834/jig.190552

李锵，王旭，关欣. 一种结合三重注意力机制的双路径网络胸片疾病分类方法［J］. 电子与信息学报， 2023， 45（4）： 1412-1425. doi: 10.11999/JEIT220172http://dx.doi.org/10.11999/JEIT220172

LI Q， WANG X， GUAN X. A dual-path network chest film disease classification method combined with a triple attention mechanism［J］. Journal of Electronics & Information Technology， 2023， 45（4）： 1412-1425.（in Chinese）. doi: 10.11999/JEIT220172http://dx.doi.org/10.11999/JEIT220172

周涛，刘赟璨，陆惠玲，等. ResNet及其在医学图像处理领域的应用：研究进展与挑战［J］. 电子与信息学报， 2022， 44（1）： 149-167. doi: 10.11999/JEIT210914http://dx.doi.org/10.11999/JEIT210914

ZHOU T， LIU Y C， LU H L， et al. ResNet and its application to medical image processing： research progress and challenges［J］. Journal of Electronics & Information Technology， 2022， 44（1）： 149-167.（in Chinese）. doi: 10.11999/JEIT210914http://dx.doi.org/10.11999/JEIT210914

HU J， SHEN L， SUN G. Squeeze-and-excitation networks［C］//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. June 18-23， 2018. Salt Lake City， UT. IEEE， 2018： 7132-7141. doi: 10.1109/cvpr.2018.00745http://dx.doi.org/10.1109/cvpr.2018.00745

范丽丽，赵宏伟，赵浩宇，等. 基于深度卷积神经网络的目标检测研究综述［J］. 光学精密工程， 2020， 28（5）： 1152-1164.

FAN L L， ZHAO H W， ZHAO H Y， et al. Survey of target detection based on deep convolutional neural networks［J］. Optics and Precision Engineering， 2020， 28（5）： 1152-1164.（in Chinese）

CHOWDHURY M E H， RAHMAN T， KHANDAKAR A， et al. Can AI help in screening viral and COVID-19 pneumonia？［J］. IEEE Access， 2020， 8： 132665-132676. doi: 10.1109/access.2020.3010287http://dx.doi.org/10.1109/access.2020.3010287

RAHMAN T， KHANDAKAR A， QIBLAWEY Y， et al. Exploring the effect of image enhancement techniques on COVID-19 detection using chest X-ray images［J］. Computers in Biology and Medicine， 2021， 132： 104319. doi: 10.1016/j.compbiomed.2021.104319http://dx.doi.org/10.1016/j.compbiomed.2021.104319

GAO S H， CHENG M M， ZHAO K， et al. Res2Net： a new multi-scale backbone architecture［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2021， 43（2）： 652-662. doi: 10.1109/tpami.2019.2938758http://dx.doi.org/10.1109/tpami.2019.2938758

HUANG G， LIU Z， VAN DER MAATEN L， et al. Densely connected convolutional networks［C］//2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. July 21-26， 2017. Honolulu， HI. IEEE， 2017： 2261-2269. doi: 10.1109/cvpr.2017.243http://dx.doi.org/10.1109/cvpr.2017.243

XIE S N， GIRSHICK R， DOLLAR P， et al. Aggregated residual transformations for deep neural networks［C］//2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. July 21-26， 2017. Honolulu， HI. IEEE， 2017： 1492-1500. doi: 10.1109/cvpr.2017.634http://dx.doi.org/10.1109/cvpr.2017.634

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. June 18-23， 2018. Salt Lake City， UT. IEEE， 2018： 4510-4520. doi: 10.1109/cvpr.2018.00474http://dx.doi.org/10.1109/cvpr.2018.00474

CHEN Y P， LI J N， XIAO H X， et al. Dual path networks［C］//Proceedings of the 31st International Conference on Neural Information Processing Systems. December 4 - 9， 2017， Long Beach， California， USA. ACM， 2017： 4470–4478.

LIU Z， LIN Y T， CAO Y， et al. Swin Transformer： hierarchical Vision Transformer using Shifted Windows［C］//2021 IEEE/CVF International Conference on Computer Vision （ICCV）. October 10-17， 2021. Montreal， QC， Canada. IEEE， 2021： 10012-10022. doi: 10.1109/iccv48922.2021.00986http://dx.doi.org/10.1109/iccv48922.2021.00986

ALAM N， KOLAWOLE S， SETHI S， et al. Vision transformers for mobile applications： a short survey［J］. arXiv preprint arXiv：2305.19365， 2023

ZHANG H， WU C R， ZHANG Z Y， et al. ResNeSt： split-attention networks［C］//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops （CVPRW）. June 19-20， 2022. New Orleans， LA， USA. IEEE， 2022： 2736-2746. doi: 10.1109/cvprw56347.2022.00309http://dx.doi.org/10.1109/cvprw56347.2022.00309

LIU Zhuang， MAO Hanzi， WU Chaoyuan， et al. A ConvNet for the 2020s［J］. arXiv preprint arXiv： 2201.03545， 2022. doi: 10.1109/cvpr52688.2022.01167http://dx.doi.org/10.1109/cvpr52688.2022.01167

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

非成对训练样本条件下的红外图像生成

基于Swin Transformer轻量化的TFT-LCD面板缺陷分类算法

全局和局部特征融合的图像去雾

跨尺度跨维度的自适应Transformer网络应用于结直肠息肉分割

面向高光谱医学图像分类的空-谱自注意力Transformer