Chest X-Ray Analysis of Tuberculosis by Deep Learning with Segmentation and Augmentation

被引:0
作者
Stirenko, Sergii [1 ]
Kochura, Yuriy [1 ]
Alienin, Oleg [1 ]
Rokovyi, Oleksandr [1 ]
Gordienko, Yuri [1 ]
Gang, Peng [2 ]
Zeng, Wei [2 ]
机构
[1] Natl Tech Univ Ukraine, Igor Sikorsky Kyiv Polytech Inst, Kiev, Ukraine
[2] Huizhou Univ, Huizhou City, Peoples R China
来源
2018 IEEE 38TH INTERNATIONAL CONFERENCE ON ELECTRONICS AND NANOTECHNOLOGY (ELNANO) | 2018年
关键词
deep learning; convolutional neural network; segmentation; open dataset; mask; data augmentation; TensorFlow; chest X-ray; computer-aided diagnosis; lung; tuberculosis; IMAGE DATABASE;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The results of chest X-ray (CXR) analysis of 2D images to get the statistically reliable predictions (availability of tuberculosis) by computer-aided diagnosis (CADx) on the basis of deep learning are presented. They demonstrate the efficiency of lung segmentation, lossless and lossy data augmentation for CADx of tuberculosis by deep convolutional neural network (CNN) applied to the small and not well-balanced dataset even. CNN demonstrates ability to train (despite overfitting) on the pre-processed dataset obtained after lung segmentation in contrast to the original not-segmented dataset. Lossless data augmentation of the segmented dataset leads to the lowest validation loss (without overfitting) and nearly the same accuracy (within the limits of standard deviation) in comparison to the original and other pre-processed datasets after lossy data augmentation. The additional limited lossy data augmentation results in the lower validation loss, but with a decrease of the validation accuracy. In conclusion, besides the more complex deep CNNs and bigger datasets, the better progress of CADx for the small and not well-balanced datasets even could be obtained by better segmentation, data augmentation, dataset stratification, and exclusion of non-evident outliers.
引用
收藏
页码:422 / 428
页数:7
相关论文
共 19 条
  • [11] Automatic Tuberculosis Screening Using Chest Radiographs
    Jaeger, Stefan
    Karargyris, Alexandros
    Candemir, Sema
    Folio, Les
    Siegelman, Jenifer
    Callaghan, Fiona
    Xue, Zhiyun
    Palaniappan, Kannappan
    Singh, Rahul K.
    Antani, Sameer
    Thoma, George
    Wang, Yi-Xiang
    Lu, Pu-Xuan
    McDonald, Clement J.
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2014, 33 (02) : 233 - 245
  • [12] KOCHURA Y, 2017, C COMP SCI INF TECHN, P243, DOI DOI 10.1007/978-3-319-70581-1_17
  • [13] Kochura Yu., 2018, DATA AUGMENTATION SE
  • [14] Deep learning
    LeCun, Yann
    Bengio, Yoshua
    Hinton, Geoffrey
    [J]. NATURE, 2015, 521 (7553) : 436 - 444
  • [15] Rajpurkar P, 2017, Arxiv, DOI [arXiv:1711.05225, DOI 10.48550/ARXIV.1711.05225]
  • [16] Shen DG, 2017, ANNU REV BIOMED ENG, V19, P221, DOI [10.1146/annurev-bioeng-071516-044442, 10.1146/annurev-bioeng-071516044442]
  • [17] Development of a digital image database for chest radiographs with and without a lung nodule: Receiver operating characteristic analysis of radiologists' detection of pulmonary nodules
    Shiraishi, J
    Katsuragawa, S
    Ikezoe, J
    Matsumoto, T
    Kobayashi, T
    Komatsu, K
    Matsui, M
    Fujita, H
    Kodera, Y
    Doi, K
    [J]. AMERICAN JOURNAL OF ROENTGENOLOGY, 2000, 174 (01) : 71 - 74
  • [18] Medical image segmentation on GPUs - A comprehensive review
    Smistad, Erik
    Falch, Thomas L.
    Bozorgi, Mohammadmehdi
    Elster, Anne C.
    Lindseth, Frank
    [J]. MEDICAL IMAGE ANALYSIS, 2015, 20 (01) : 1 - 18
  • [19] Spatiotemporal Pyramid Network for Video Action Recognition
    Wang, Yunbo
    Long, Mingsheng
    Wang, Jianmin
    Yu, Philip S.
    [J]. 30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 2097 - 2106