基于密集卷积网络的X线气胸检测与定位

doi:10.11772/j.issn.1001-9081.2019050884

计算机应用 ›› 2019, Vol. 39 ›› Issue (12): 3541-3547.DOI: 10.11772/j.issn.1001-9081.2019050884

基于密集卷积网络的X线气胸检测与定位

罗国婷¹, 刘志勤¹, 周莹², 王庆凤¹, 郑介志³, 刘启榆²

1. 西南科技大学计算机科学与技术学院, 四川绵阳 621000;
2. 绵阳市中心医院放射科, 四川绵阳 621000;
3. 上海联影智能医疗科技有限公司, 上海 200232

收稿日期:2019-05-27 修回日期:2019-07-23 出版日期:2019-12-10 发布日期:2019-09-10
作者简介:罗国婷(1995-),女,四川绵阳人,硕士研究生,CCF会员,主要研究方向:医学图像分析、深度学习;刘志勤(1962-),女,四川绵阳人,教授,硕士,CCF会员,主要研究方向:医学图像分析、高性能计算;周莹(1984-),女,四川绵阳人,主治医师,博士,主要研究方向:肿瘤图像分析、深度学习;王庆凤(1988-),女,四川安岳人,讲师,博士,CCF会员,主要研究方向:医学图像分析、计算机辅助诊断;郑介志(1980-),男,中国台湾人,博士,主要研究方向:医学图像分析、模式识别;刘启榆(1963-),男,四川广元人,主任医师,硕士,主要研究方向:医学图像分析、介入放射学。
基金资助:
四川省科技计划项目（2019JDRC0119）。

Pneumothorax detection and localization in X-ray images based on dense convolutional network

LUO Guoting¹, LIU Zhiqin¹, ZHOU Ying², WANG Qingfeng¹, CHENG Jiezhi³, LIU Qiyu²

1. College of Computer Science and Technology, Southwest University of Science and Technology, Mianyang Sichuan 621000, China;
2. Radiology Department, Mianyang Central Hospital, Mianyang Sichuan 621000, China;
3. Shanghai United Imaging Intelligence Limited Company, Shanghai 200232, China

Received:2019-05-27 Revised:2019-07-23 Online:2019-12-10 Published:2019-09-10
Contact: 刘志勤
Supported by:
This work is partially supported by the Science and Technology Program of Sichuan Province (2019JDRC0119).

摘要/Abstract

摘要： 现有X线气胸检测存在两个主要问题：一是由于气胸通常与肋骨、锁骨等组织重叠，在临床上存在较大的漏诊，而现有算法的检测性能仍有待提高；二是现有基于卷积神经网络的算法无法给出可疑的气胸区域，缺乏可解释性。针对上述问题，提出了一种结合密集卷积网络（DenseNet）与梯度加权类激活映射的方法用于X线气胸的检测与定位。首先，构建了一个较大规模的胸部X线数据集PX-ray用于模型的训练和测试。其次，修改DenseNet的输出节点并在全连接层后添加一个sigmoid函数对胸片进行二分类（气胸/非气胸）。在训练过程中通过设置交叉熵损失函数的权重来缓解数据不平衡问题，提高模型准确率。最后，提取网络最后一个卷积层的参数以及对应的梯度，通过梯度加权类激活映射算法获得气胸类别的粗略定位图。在PX-ray测试集上的实验结果表明，所提方法的检测准确率为95.45%，并且在曲线下面积（AUC）、敏感度、特异性等指标上均高于0.9，优于VGG19、GoogLeNet以及ResNet算法，同时实现了对气胸区域的可视化。

关键词: 气胸, 胸部X线, 密集卷积网络, 类别不平衡, 类激活映射

Abstract: There are two main problems about pneumothorax detection in X-ray images. The pneumothorax usually overlaps with tissues such as ribs and clavicles in X-ray images, easily causing missed diagnosis and the performance of the existing pneumothorax detection methods remain to be improved. The suspicious pneumothorax area detection cannot be exploited by the convolutional neural network-based algorithms, lacking the interpretability. Aiming at the problems, a novel method combining Dense convolutional Network (DenseNet) and gradient-weighted class activation mapping was proposed. Firstly, a large-scale chest X-ray dataset named PX-ray was constructed for model training and testing. Secondly, the output node of the DenseNet was modified and a sigmoid function was added after the fully connected layer to classify the chest X-ray images. In the training process, the weight of cross entropy loss function was set to alleviate the problem of data imbalance and improve the accuracy of the model. Finally, the parameters of the last convolutional layer of the network and the corresponding gradients were extracted, and the areas of the pneumothorax type were roughly located by gradient-weighted class activation mapping. The experimental results show that, the proposed method has the detection accuracy of 95.45%, and has the indicators such as Area Under Curve (AUC), sensitivity, specificity all higher than 0.9, performs the classic algorithms of VGG19, GoogLeNet and ResNet, and realizes the visualization of pneumothorax area.

Key words: pneumothorax, chest X-ray, Dense convolutional Network (DenseNet), class-imbalance, class activation mapping

中图分类号:

TP391.4

罗国婷, 刘志勤, 周莹, 王庆凤, 郑介志, 刘启榆. 基于密集卷积网络的X线气胸检测与定位[J]. 计算机应用, 2019, 39(12): 3541-3547.

LUO Guoting, LIU Zhiqin, ZHOU Ying, WANG Qingfeng, CHENG Jiezhi, LIU Qiyu. Pneumothorax detection and localization in X-ray images based on dense convolutional network[J]. Journal of Computer Applications, 2019, 39(12): 3541-3547.

参考文献

[1] 王朝,邹卫.原发性自发性气胸病因研究进展[J].临床肺科杂志,2015,20(6):1120-1122,1126.(WANG C, ZOU W. Progress in study of primary spontaneous pneumothorax[J]. Journal of Clinical Pulmonary Medicine, 2015, 20(6):1120-1122, 1126.)
[2] THOMSEN L, NATHO O, FEIGEN U, et al. Value of digital radiography in expiration in detection of pneumothorax[J]. RoFo:Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin, 2014, 186(3):267-273.
[3] BINTCLIFFE O, MASKELL N. Spontaneous pneumothorax[J]. British Medical Journal, 2014, 348:g2928.
[4] SUTHAR M, MAHJOUBFAR A, SEALS K, et al. Diagnostic tool for pneumothorax[C]//Proceedings of the 2016 IEEE Photonics Society Summer Topical Meeting Series. Piscataway:IEEE, 2016:218-219.
[5] MACDUFF A, ARNOLD A, HARVEY J. Management of spontaneous pneumothorax:British thoracic society pleural disease guideline 2010[J]. Thorax, 2010, 65(S2):ii18-ii31.
[6] O'CONNOR A R, MORGAN W E. Radiological review of pneumothrax[J]. British Medical Journal, 2005, 330(7506):1493-1497.
[7] SEOW A, KAZEROONI E A, PERNICANO P G, et al. Comparison of upright inspiratory and expiratory chest radiographs for detecting pneumothoraces[J]. American Journal of Roentgenology, 1996, 166(2):313-316.
[8] 穆魁津.谈谈气胸的诊断方法[J].中国实用内科杂志,1995,15(4):205-206.(MU K J. Discussion on the diagnosis of pneumothorax[J]. Chinese Journal of Practical Internal Medicine, 1995, 15(4):205-206).
[9] SANADA S, DOI K, MACMAHON H. Image feature analysis and computer-aided diagnosis in digital radiography:automated detection of pneumothorax in chest images[J]. Medical Physics, 1992, 19(5):1153-1160.
[10] GEVA O, ZIMMERMAN-MORENO G, LIEBERMAN S, et al. Pneumothorax detection in chest radiographs using local and global texture signatures[C]//Proceedings of the Medical Imaging 2015:Computer-Aided Diagnosis, SPIE 9414. Bellingham:SPIE, 2015:Article No. 94141P.
[11] CHAN Y H, ZENG Y Z, WU H C, et al. Effective pneumothorax detection for chest X-ray images using local binary pattern and support vector machine[J]. Journal of Healthcare Engineering, 2018, 2018:Article No. 2908517.
[12] LECUN Y, BOSER B, DENKER J S, et al. Handwritten digit recognition with a back-propagation network[C]//Proceedings of the 1989 International Conference on Neural Information Processing Systems. San Francisco:Morgan Kaufmann Publishers Inc., 1989:396-404.
[13] 何雪英,韩忠义,魏本征.基于深度卷积神经网络的色素性皮肤病识别分类[J].计算机应用,2018,38(11):3236-3240.(HE X Y, HAN Z Y, WEI B Z. Pigmented skin lesion classification based dermoscopic images using deep convolutional neural network[J]. Journal of Computer Applications, 2018, 38(11):3236-3240).
[14] WANG X, PENG Y, LU L, et al. Chestx-ray8:hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:3462-3471.
[15] JUN T J, KIM D, KIM D. Automated diagnosis of pneumothorax using an ensemble of convolutional neural networks with multi-sized chest radiography images[EB/OL].[2019-03-22]. https://arxiv.org/ftp/arxiv/papers/1804/1804.06821.pdf.
[16] TAYLOR A G, MIELKE C, MONGAN J. Automated detection of moderate and large pneumothorax on frontal chest X-rays using deep convolutional neural networks:a retrospective study[J]. PLOS Medicine, 2018, 15(11):e1002697.
[17] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2015:1-9.
[18] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2019-03-22]. https://arxiv.org/pdf/1409.1556.pdf.
[19] PINHEIRO P O, COLLOBERT R. From image-level to pixel-level labeling with convolutional networks[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2015:1713-1721.
[20] RIBEIRO M T, SINGH S, GUESTRIN C. "Why Should I Trust You?":explaining the predictions of any classifier[EB/OL].[2019-03-22]. https://arxiv.org/pdf/1602.04938.pdf.
[21] FANG H, GUPTA S, IANDOLA F, et al. From captions to visual concepts and back[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2015:1473-1482.
[22] ZHOU B, KHOSLA A, LAPEDRIZA A, et al. Learning deep features for discriminative localization[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2016:2921-2929.
[23] SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM:visual explanations from deep networks via gradient-based localization[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway:IEEE, 2017:618-626.
[24] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:2261-2269.
[25] BENGIO Y, COURVILLE A, VINCENT P. Representation learning:a review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8):1798-1828.
[26] MAHENDRAN A, VEDALDI A. Visualizing deep convolutional neural networks using natural pre-images[J]. International Journal of Computer Vision, 2016, 120(3):233-255.
[27] 杨真真,匡楠,范露,等.基于卷积神经网络的图像分类算法综述[J].信号处理,2018,34(12):1474-1489.(YANG Z Z, KUANG N, FAN L, et al. Review of image classification algorithms based on convolutional neural networks[J]. Journal of Signal Processing, 2018, 34(12):1474-1489.)

基于密集卷积网络的X线气胸检测与定位

Pneumothorax detection and localization in X-ray images based on dense convolutional network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

[1]	何雪英, 韩忠义, 魏本征. 基于深度卷积神经网络的色素性皮肤病识别分类[J]. 计算机应用, 2018, 38(11): 3236-3240.
[2]	郝宁, 夏士雄, 牛强, 赵志军. 基于类别重要度的MIMLBoost改进算法[J]. 计算机应用, 2015, 35(11): 3122-3125.