臺灣博碩士論文加值系統

本論文提出一個基於卷積神經網路架構(CNNs)的系統，並用於自動分類並辨識顯微鏡觀測人類幹細胞區域影像。我們的方法能偵測出reprogramming和reprogrammed的人類誘導性多功能幹細胞，使生物學家能進一步以試劑篩查或條件培養達成幹細胞培養。學習結果展示出我們的CNNs可以已top-1 error rates 為5.9%和top-2 error rates 為0.9%的準確度繪製機率分布圖並用以分析。實作結果顯示出這個自動化方法可以成功偵測出並定位出人類幹細胞組成，從而得到一個對幹細胞研究有幫助的潛力工具。

We present a deep learning architecture Convolutional Neural Networks (CNNs) for automatic classification and recognition of reprogramming and reprogrammed human Induced Pluripotent Stem (iPS) cell regions in microscopy images. The differentiated cells that possibly undergo reprogramming to iPS cells can be detected by this method for screening reagents or culture conditions in iPS induction. The learning results demonstrate that our CNNs can achieve the Top-1 and Top-2 error rates of 5.9% and 0.9%, respectively, to produce probability maps for the automatic analysis. The implementation results show that this automatic method can successfully detect and localize the human iPS cell formation, thereby yield a potential tool for helping iPS cell culture.

目錄
摘要 I
Abstract II
誌謝 III
目錄 IV
圖索引 VI
表索引 VII
第一章 緒論 1
1.1 研究背景 1
1.2 相關研究 2
1.3 研究動機與目的 4
1.4 論文架構 6
第二章 系統架構 7
2.1 Caffe ─ 一套完整且易於使用的CNN工具 9
2.2卷積神經網路模型 14
2.3參數設置 17
第三章 模組訓練 18
3.1訓練影像分類與切割 18
3.2模組訓練準備與過程 21
第四章 視覺化與正規化 24
4.1 機率圖產生與視覺化 25
4.2 結果分析 27
第五章 實例─人類iPS細胞定位與辨識 28
5.1 iPS相關資訊 29
5.1.1 iPS細胞培養 29
5.1.2 成像條件 31
5.2 細胞分類 33
5.3 細胞影像切割 35
5.4 模組訓練結果 38
5.5 研究結果 39
5.5.1 線索一：類別四+五與類別六 40
5.5.2 線索二：類別一與類別六 44
第六章 結論與未來展望 48
參考文獻 49

圖索引
圖2-1 本研究系統流程圖 7
圖2-2 特徵提取層示意圖 13
圖2-3 bvlc reference caffenet架構 16
圖3-1 自動裁切工具輸出的檔案 18
圖3-2 自動裁切工具的使用範例 19
圖3-3 模組訓練過程 23
圖4-1 視覺化與正規化的流程圖 24
圖4-2 機率圖範例與原圖 26
圖4-3 草地範圍 27
圖5-1 人類iPS細胞顯微鏡影像範例 32
圖5-2 六類細胞範例 34
圖5-3 各類別裁切結果範例 36
圖5-4 第六類別：iPS細胞 36
圖5-5人類iPS細胞 37
圖5-6 IWP-2 Colony 6類別四+五與類別六的機率圖比較 36
圖5-7 IWP-2 Colony 2類別四+五與類別六的機率圖比較 37
圖5-8 IWP-2 Colony 6類別一與類別六的機率圖比較 45
圖5-9 IWP-2 Colony 2類別一與類別六的機率圖比較 46

表索引
表5-1系統研究環境 28
表5-2模組實測分類結果 38
表5-3 IWP-2 Colony 6類別四+五與類別六的折線比較圖 43
表5-4 IWP-2 Colony 2類別四+五與類別六的折線比較圖 43
表5-5 IWP-2 Colony 6類別一與類別六的折線比較圖 47
表5-6 IWP-2 Colony 2類別一與類別六的折線比較圖 47

參考文獻
[1].H. Klassen, “Stem cells in clinical trials for treatment of retinal degeneration,” Expert Opinion on Biological Therapy, vol. 16(1), pp. 7–14, 2016.
[2].H. Klassen, “Stem cells in clinical trials for treatment of retinal degeneration,” Expert Opinion on Biological Therapy, vol. 16(1), pp. 7–14, 2016.
[3].LeCun, Yann, et al. “Gradient-based learning applied to document recognition.” Proceedings of the IEEE 86.11 (1998): 2278-2324.
[4].THE MNIST DATABASE of handwritten digits, Yann LeCun, Courant Institute, NYU, orinna Cortes, Google Labs, New York, Christopher J.C. Burges, Microsoft Research, Redmond
[5].Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich. Going deeper with convolutions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.1–9, 2015.
[6].Large Scale Visual Recognition Challenge 2014 (ILSVRC2014) . Olga Russakovsky, Stanford University. Sean Ma, Stanford University. Jonathan Krause, Stanford University. Jia Deng, University of Michigan. Alex Berg, UNC Chapel Hill. Fei-Fei Li, Stanford University.
[7].D. C. Cireşan, A. Giusti, L. M. Gambardella, J. Schmidhuber. “Mitosis Detection in Breast Cancer Histology Images with Deep Neural Networks” MICCAI 2013.
[8].S. B. Lo, S. A. Lou, J. Lin, M. T. Freedman, M. V. Chien, S. K. Mun. “Artificial Convolution Neural Network Techniques and Applications for Lung Nodule Detection.” IEEE Trans. Medical imaging, vol. 14, no. 4, December 1995.
[9].Y. Jia , E. Shelhamer , J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, T. Darrell, “Caffe: Convolutional Architecture for Fast Feature Embedding” arXiv preprint arXiv:1408.5093, 2014.
[10].深度學習 : 21天實戰 Caffe. 趙永科著
[11].G. Bradski, A. Kaehler, Learning OpenCV, Massachusetts: O’Reilly Media, 2008.
[12].H. Klassen, “Stem cells in clinical trials for treatment of retinal degeneration,” Expert Opinion on Biological Therapy, vol. 16(1), pp. 7–14, 2016.
[13]. H. Klassen, “Stem cells in clinical trials for treatment of retinal degeneration,” Expert Opinion on Biological Therapy, vol. 16(1), pp. 7–14, 2016.