使用卷積神經網路辨別胸腔X光片內異常

Identify Chest X-ray Abnormalities by using Convolutional Neural Network

王立昀(Li-Yun Wang)

指導教授 ： 古政元 ； 郭文嘉

國立交通大學/管理學院/資訊管理研究所/碩士(2019年)

摘要

本論文中詳細介紹卷積神經網路(Convolution Neural Network, CNN)應用於圖形辨識的技術，包含其歷史演進、基本架構、研究中常用模型，並透過簡單的舉例，使在醫院工作的非資訊人員，在閱讀完本論文後，能對CNN有基礎的認知。同時，本論文以Rajpurkar et al.所提出，名為CheXNet的CNN架構[1]為基礎，提出一個新的分類方法，建構一個能夠辨別胸腔X光片中14種異常的分類辨識模型。最終將影像分為14種不同的異常：肺不張(Atelectasis)、心臟肥大(Cardiomegaly)、肺積液(Effusion)、浸潤(Infiltration)、腫塊(Mass)、結節(Nodule)、肺炎(Pneumonia)、氣胸(Pneumothorax)、肺實變(Consolidation)、肺水腫(Edema)、肺氣腫(Emphysema)、纖維化(Fibrosis)、胸膜增厚(Pleural Thickening)、疝氣(Hernia)。期望此模型能夠達到高辨識率，輔助年輕醫師進行診斷以及補足偏鄉地區醫師資源不足的問題，提供人們平等的醫療檢查。

關鍵字

卷積神經網路 ； 胸腔X光片 ； 電腦輔助診斷

並列摘要

In this paper, we will introduce the Convolutional Neural Network (CNN) technology for graphic recognition in detail. Include its history, basic structure, commonly used models in research, and through simple examples to make non-information background personnel working in hospitals can have a basic understanding of CNN after reading this paper. In addition, based on the CNN architecture proposed by Rajpurkar et al., this paper proposes a new classification method to construct a classification identification model that can identify 14 abnormalities in chest X-ray films[1]. The images were eventually divided into 14 different conditions: Atelectasis, Cardiomegaly, Effusion, Infiltration, Mass, Nodule, Pneumonia, Pneumothorax, Consolidation, Edema, Emphysema, Fibrosis, Pleural Thickening, Hernia. Hope to achieve high recognition rate, assist young doctors in diagnosis and make up for the shortage of doctor resources in the rural areas, and provide equal medical examinations.

並列關鍵字

CNN ； Chest X-ray ； CAD

參考文獻

1. Rajpurkar, P., Irvin, J., Zhu, K., Yang, B., Mehta, H., Duan, T., Ding, D., Bagul, A., Ball, R., Langlotz, C., Shpanskaya, K., Lungren, M., Ng, A. (2017).CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning.

2. Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M., & Summers, R. M. (2017). Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2097-2106.

3. Cicero, M., Bilbily, A., Colak, E., Dowdell, T., Gray, B., Perampaladas, K., & Barfett, J. 2017. Training and validating a deep convolutional neural network for computer-aided detection and classification of abnormalities on frontal chest radiographs. Investigative radiology, 52(5), 281-287.

4. LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.

5. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, 1097-1105.