间质纤维化CT衰减模式的整体分类通过深度卷积神经网络进行肺部疾病
Overall Classification of Interstitial Fibrosis CT Attenuation Patterns for Pulmonary Disease through Deep Convolutional Neural Networks
摘要: 间质性肺病(ILD)涉及在计算机断层扫描(CT)图像中观察到的几种异常成像模式。这些模式的准确分类在其中起着重要作用。准确地检测疾病的性质和发展程度。因此，开发自动肺部计算机辅助检测系统非常重要。传统上，该任务依赖于专家手动识别感兴趣区域(ROI)作为诊断潜在疾病的先决条件。该协议耗时并且禁止全自动评估。在这本文提出了一种在CT图像上对ILD成像模式进行分类的新方法。主要区别在于所提出的算法使用整个图像作为整体输入。通过规避手动输入ROI的先决条件，我们所面临的比以前更困难，但可以更好地解决临床工作流程。使用公开的ILD数据库的定性和定量结果证明了基于补丁的分类下的最新分类准确性，并显示了使用整体图像预测ILD类型的潜力。
Abstract: Interstitial lung disease (ILD) involves several abnormal imaging patterns observed in computed tomography (CT) images. The accurate classification of these models plays an important role in it. Clinical testing can effectively diagnose the nature of the disease. Therefore, it is very important to develop an automated lung computer-aided detection system. Traditionally, this task relies on experts manually identifying regions of interest (ROI) as a prerequisite for diagnosing underlying disease. This protocol is time consuming and does not allow for fully automated evaluation. In this paper, a new method for classifying ILD imaging modes on CT images is proposed. The main dif-ference is that the proposed algorithm uses the entire image as a whole input. By circumventing the prerequisites for manually entering ROI, our problem setting is more significantly. It is more difficult than previous work, but it can better solve the clinical workflow. The qualitative and quantitative results using the published ILD database demonstrate the latest classification accuracy under the patch-based classification and show the potential to predict the ILD type using the overall image.
文章引用：于红梅, 武建强. 间质纤维化CT衰减模式的整体分类通过深度卷积神经网络进行肺部疾病[J]. 医学诊断, 2019, 9(2): 61-65. https://doi.org/10.12677/MD.2019.92012

参考文献

[1] Bağcı, U., Bray, M., Caban, J., Yao, J. and Mollura, D.J. (2012) Computer-Assisted Detection of Infectious Lung Diseases: A Review. Computerized Medical Imaging and Graphics, 36, 72-84. [Google Scholar] [CrossRef] [PubMed]
[2] Bagci, U., Yao, J., Wu, A., Caban, J., Palmore, T., Suffredini, A., Aras, O. and Mollura, D. (2012) Automatic Detection and Quantification of Tree-in-Bud (TIB) Opacities from CT Scans. IEEE Transactions on Biomedical Engineering, 59, 1620-1632. [Google Scholar] [CrossRef] 
[3] Cireşan, D.C., Giusti, A., Gambardella, L.M. and Schmidhuber, J. (2013) Mitosis Detection in Breast Cancer Histology Images with Deep Neural Networks. International Conference on Medical Image Computing and Computer-Assisted Intervention, Nagoya, 22-26 September 2013, 411-418.
[4] Li, G., Wang, L., Shi, F., et al. (2013) Multi-Atlas Based Simultaneous Labeling of Longitudinal Dynamic Cortical Surfaces in Infants. Med Image Comput Comput Assist Interv, 16, 58-65.
[5] Gao, M., Bagci, U., Lu, L., et al. (2015) Holistic Classification of CT Attenuation Patterns for Interstitial Lung Diseases via Deep Convolutional Neural Networks. Computer Methods in Biomechanics and Biomedical Engineering. Imaging & Visualization, 6, 1-6. [Google Scholar] [CrossRef] [PubMed]
[6] Leynes, A.P., Yang, J., Shanbhag, D.D., et al. (2017) Hybrid ZTE/Dixon MR-Based Attenuation Correction for Quantitative Uptake Estimation of Pelvic Lesions in PET/MRI. Medical Physics, 44, 902-913.
[7] Helms, C.A. and Wall, S.D. (1985) CT Attenuation Numbers in the Lumbar Spine and Their Utility in Diagnosing Disc Disease. Computerized Radiology, 9, 291-297. [Google Scholar] [CrossRef] [PubMed]
[8] Bowen, S.R., Nyflot, M.J., Zeng, J., et al. (2013) TU-E-141-09: Impact of Attenuation Correction Mode on 4D PET/CT for Target Definition in Lung Cancer Patients. Medical Physics, 40, 449-449. [Google Scholar] [CrossRef] 
[9] Alshehri, S.M., Naushad, M., Ahamad, T., et al. (2014) Synthesis, Characterization of Curcumin Based Ecofriendly Antimicrobial Bio-Adsorbent for the Removal of Phenol from Aqueous Medium. Chemical Engineering Journal, 254, 181-189.
[10] Gessert, N., Heyder, M., Latus, S., et al. (2018) Plaque Classification in Coronary Arteries from IVOCT Images Using Convolutional Neural Networks and Transfer Learning. The International Journal for Computer Assisted Radiology and Surgery, 13, S99-S100.
[11] Xue, M., Tang, Y., Wu, L., et al. (2018) Model Approximation for Switched Genetic Regulatory Networks. IEEE Transactions on Neural Networks & Learning Systems, 29, 3404-3417. [Google Scholar] [CrossRef] 
[12] Melrose, R.J., Jimenez, A.M., Riskin-Jones, H., et al. (2018) Alterations to Task Positive and Task Negative Networks during Executive Functioning in Mild Cognitive Impairment. Neuroimage Clinical, 19, 970-981. [Google Scholar] [CrossRef] [PubMed]
[13] Qu, Y., Fang, B., Zhang, W., et al. (2018) Product-Based Neural Networks for User Response Prediction over Multi-Field Categorical Data. ACM Transactions on Information Systems, 37, Article No. 5. [Google Scholar] [CrossRef] 
[14] Bölcskei, H., Grohs, P., Kutyniok, G., et al. (2018) Optimal Approximation with Sparsely Connected Deep Neural Networks. SIAM Journal on Mathematics of Data Science, 1, 8-45. [Google Scholar] [CrossRef] 
[15] Parisi, G.I., Kemker, R., Part, J.L., et al. (2018) Continual Lifelong Learning with Neural Networks: A Review. Neural Networks, 113, 54-71. [Google Scholar] [CrossRef] [PubMed]
[16] Wu, S., Li, G., Chen, F., et al. (2018) Training and Inference with Integers in Deep Neural Networks.
[17] Li, R., Wang, S., Zhu, F., et al. (2018) Adaptive Graph Convolutional Neural Networks.