間質性肺病(Interstitial lung disease, ILD)，又稱為瀰漫性肺病，會造成肺泡的結構改變，並且波及肺泡的實質成分與氣管的分支處，初期症狀為咳嗽、胸悶，嚴重時則會導致呼吸功能衰竭、肺萎縮等，間質性肺病包含了200多個影響肺間質的病種，儘管每種疾病的臨床表現有各自的特徵，不同類的病種仍然有很多共通點。此外，有些病種就算得到了切片，醫師還是不能準確判斷疾病的種類，醫師只能結合經驗與參考切片來提高判別率，更別說是實習醫師了，近年有許多特徵提取的方法被提出，但是大部分只能應用在特定領域。本論文分別設計了兩種較淺的卷積網路模型，第一種適用於間質性肺病的二分類，第二種適用於間質性肺病的五分類，透過正則化方法與不同的超參數配置，得到兩個較淺架構的卷積神經網路模型，在避免過度擬合問題的同時，還能有不錯的分類準確率，此外，這兩種模型還能推廣應用到紋理特徵的任務中。實驗結果顯示，本論文提出的模型可以用於醫師的輔助診斷。

Interstitial lung disease (ILD), which is also referred to as diffuse parenchymal lung disease (DPLD), causes structural changes in the alveoli. In addition, it affects the actual components of the alveoli and the branches of the trachea. The initial symptoms are a cough and chest tightness. In severe cases, it results in Respiratory failure, lung atrophy, etc. Interstitial lung disease contains more than 200 diseases affecting the lung parenchyma. Although the clinical manifestations of each disease have their own characteristics, there is still a lot in common between different types of diseases. In addition, even if some diseases have been sliced, doctors still cannot accurately determine the type of disease. Doctors can only improve the discriminant rate by combining experience and refer to slices, not to mention intern doctors. Over the past few years, many feature extraction methods have been proposed, but most of them can only be applied in specific fields. In this thesis, two shallow varieties of convolutional network models have been designed. The first one is applicable to the two categories classification of interstitial lung disease while the second is applicable to the five categories classification of interstitial lung diseases. With regularization method and different hyperparameters configuration, models are able to accurately classify images and simultaneously avoid from overfitting. In addition, these two models can be adapted to texture-like features task. Experimental results show that the model presented in this thesis can be used for assistant diagnosis of doctors.

REFERENCE
[1] T. E. King. Jr, “Clinical advances in the diagnosis and therapy of the interstitial lung disease,” Am J Respir Crit Care Med, vol. 172, no. 3, 2005
[2] M. Ng, T. Fleming, M. Robinson, et al. “Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990-2013: a systematic analysis for the global burden of disease study 2013,” Lancet, 2015
[3] M. Naghavi, H. Wang, R. Lozano, et al. “Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the global burden of disease study 2013,” Lacent, 2015
[4] “Golgi Staining Technique,” http://embryo.asu.edu/pages/golgi-staining-technique
[5] C. Golgi, “The neuron doctrine—theory and facts,” Nobel Lecture, 1906
[6] H. Sompolinsky, “The theory of neural networks: The Hebb rules and beyond,” Springer Lecture Notes in Physics, vol. 275, pp. 485-527, 1987
[7] F. Rosenblatt, “The perceptron: a probabilistic model for information storage and organization in the brain,” Psychological Review, vol. 65, no. 6, pp. 386-408, 1958
[8] D. H. Ackley, G. E. Hinton and T. J. Sejnowski, “A learning algorithm for boltzmann machines,” Cognitive Science, vol. 9, no. 1, pp. 147-169, 1985
[9] D. E. Rumelhart, G. E. Hinton and R. J. Williams, “Learning representations by back-propagating errors,” Nature, vol. 323, pp. 533-536, 1986
[10] Y. Lecun, L. Bottou, Y. Bengio and P. Haffner, “Gradient-based learning applied to dcument recognition,” Proc. IEEE, vol. 86, no. 11, pp. 2278-2324, 1998
[11] 李彥冬, 郝宗波, 雷航,“卷積神經網路研究綜述,” Journal of Computer Applications, vol. 36, no. 9, pp. 2508-2515, 2016
[12] A. Waibel, T. Hanasawa, G. Hinton, K. Shikano and K. J. Lang, “Phoneme recognition using time-delay neural networks,” IEEE Trans. Acoust, Speech and Signal Processing, vol. 37, no.3, pp. 328-339, 1989
[13] R. Vaillant, C. Monrocq and Y. L. Cun, “Orginal approach for the localisation of objects in images,” In Proc. Vision, Image and Signal Processing, vol. 141, no. 4, pp. 245-250, 1994
[14] S. Lawrence, C. L. Giles, A. C. Tsoi and A. D. Back, “Face recognition: a convolutional neural-network approach,” IEEE Trans. Neural Networks, vol. 8, no. 1, pp. 98-113, 1997
[15] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever and R. Salakhutdinov, “Dropout: a simple way to prevent neural networks from overfitting,” J. Machine Learning Research, vol. 15, pp. 1929-1958, 2014
[16] T. Salimans and D. P. Kingma, “Weight normalization: a simple reparameterization to accelerate training of deep neural networks,” Advances in Neural Information Processing Systems, pp. 901-909, 2016
[17] D. Silver, et al., “Mastering the game of go with deep neural networks and tree search,” Nature, vol. 529, pp. 484-489, 2016
[18] K. He, X. Zhang, S. Ren and J. Sun, “Deep residual learning for image recognition,” IEEE Computer Vision and Pattern Recognition, 2015
[19] “A brief introduction to artificial neural networks,” http://tuxar.uk/brief-introduction-artificial-neural-networks/
[20] “Blue mountain神經網路及其PID控制,”
http://www.cnblogs.com/BlueMountain-HaggenDazs/p/6160411.html
[21] “Chapter 1 Using neural nets to recognize handwritten digits,” http://neuralnetworksanddeeplearning.com/chap1.html
[22] “Drawing an unfolded recurrent neural network,”
https://tex.stackexchange.com/questions/364413/drawing-an-unfolded-recurrent-neural-network
[23] “Mr. Opengate,”
http://mropengate.blogspot.com/2015/05/ai-supervised-learning.html
[24] “ML Algos from Scratch-KMeans Clustering,”
http://www.datascribble.com/blog/machine-learning/ml-algos-scratch-kmeans-clustering/
[25] “Semi-Supervised Learning,”
https://www.slideshare.net/lukastencer/semisupervised-learning-42075774
[26] “Reinforcement Learning with Q tables,” https://itnext.io/reinforcement-learning-with-q-tables-5f11168862c8
[27] K. Fukushima, “Neocognitron: a self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position,” Biological Cybernetics, vol. 36, no. 4, pp. 193-202, 1980
[28] D. H. Hubel, and T. N. Wiesel, “Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex,” Journal of Physiology, vol. 160, pp. 106-154, 1962
[29] “Convolutional neural network,”
http://speech.ee.ntu.edu.tw/~tlkagk/courses/ML_2017/Lecture/CNN.pdf
[30] S. Ruder, “An overview of gradient descent optimization algorithms,” arXiv preprint, arXiv: 1609.04747, 2016
[31] “Gradient descent,”
http://speech.ee.ntu.edu.tw/~tlkagk/courses/ML_2017/Lecture/Gradient%20Descent.pdf
[32] “How the backpropagation algorithm works,” neuralnetworksanddeeplearning.com/chap2.html#proof_of_the_four_fundamental_equations_(optional)
[33] “Backpropagation,” https://itw01.com/GF2IELL.html
[34] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “imageNet classification with deep convolutional neural networks,” NIPS Proceedings, pp. 1106-1114, 2012
[35] A. Depeursinge, et al., “Building a reference multimedia database for interstitial lung diseases,” Comput. Med. Imaging and. Grap., vol. 36, no. 3, pp. 227-238, 2012