跳到主要內容

臺灣博碩士論文加值系統

(44.211.239.1) 您好!臺灣時間:2023/01/31 06:34
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:張亘亘碩
研究生(外文):CHANG,SHENG-SHUO
論文名稱:建構機器學習技術為基礎的生物組織偏 振光學特性自動辨識系統之研究
論文名稱(外文):Constructing an Automatic Identification System for the Polarization Optical Properties of Biological Tissues Based on Machine Learning Technology
指導教授:韓建遠
指導教授(外文):HAN,CHIEN-YUAN
口試委員:游智仁潘國興
口試委員(外文):YU,CHIH-JENPHAN,QUOC-HUNG
口試日期:2022-01-20
學位類別:碩士
校院名稱:國立聯合大學
系所名稱:光電工程學系碩士班
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2022
畢業學年度:110
語文別:中文
論文頁數:86
中文關鍵詞:穆勒矩陣極分解層析能力生物組織辨識深度學習
外文關鍵詞:Mueller matrixDecompositionTomography abilityBiological tissue identificationDeep learning
相關次數:
  • 被引用被引用:0
  • 點閱點閱:57
  • 評分評分:
  • 下載下載:10
  • 收藏至我的研究室書目清單書目收藏:1
本實驗使用Python語言編寫卷積神經網路(Convolutional Neural Network, CNN)建構生物組織偏振光學特性機器學習自動辨識系統。使用已經發展完成的混和調變式穆勒矩陣影像量測系統測量高分化肝細胞癌細胞與非腫瘤性肝細胞的切片組織,獲得細胞的穆勒矩陣影像及分解穆勒矩陣得到的細胞組織偏振特性影像,將其中可以觀察到差異性最明顯的偏振度影像用於深度學習訓練。最後利用迭代循環訓練及驗證獲得的模型來辨識生物細胞是否為癌變細胞,此方法相較於一般的人工辨識,僅需要使用訓練過後的模型,便能更快速且準確的對樣本進行分類,進而提升穆勒矩陣影像光學特性辨識之效率。
In this work, we constructing an automatic identification system (AIS) for the polarization optical properties of biological tissues based on machine learning technology with convolutional neural network (CNN) executed of Python language. We measure the biological tissue of hepatocellular carcinoma and non-neoplastic liver from a Muller matrix image system with hybrid modulation technique. Hybrid modulation technique has been developed for analysis the Muller matrix image of the cell, and decomposed Muller matrix to obtain the most obvious polarization images can be observed for deep learning training. Finally, the results obtained by iterative loop training and verification are used to distinguish whether biological cells are cancerous. Compared with general manual identification, this method only needs to use the trained model to classify samples more quickly and accurately, Thereby improve the efficiency of the recognition of the optical characteristics of the Muller matrix image.
誌謝 I
摘要 II
Abstract III
目錄 IV
圖目錄 VII
表目錄 X
第一章、 研究背景與目的 1
1.1研究背景 1
1.2 論文架構 5
第二章、 基本原理介紹 6
2.1 偏振光學表示方法 6
2.2 史托克參數與穆勒矩陣 8
2.2.1 偏振率 10
2.2.2穆勒矩陣 11
2.3 穆勒矩陣各項元素基本意義 13
2.3.1 衰減特性(Diattenuation effect) 14
2.3.2 偏振度(Polarizance) 15
第三章、 穆勒矩陣之極分解 18
3.1 Lu-Chipman極分解法(Polar decomposition) 18
3.1.1矩陣分解 19
3.1.2 衰減矩陣分解 20
3.1.3 退偏振能力 22
3.1.4 相位延遲係數 23
3.2 微分矩陣分解法(Differential decomposition) 25
3.2.1 微分矩陣參數意義 27
3.2.2 非退偏振與退偏振表示方法 29
3.2.3 相關光學參數解析方式 30
第四章、 深度學習系統 32
4.1神經網路與深度學習 32
4.2 卷積神經網路(Convolutional Neural Network, CNN) 40
第五章、 深度學習穆勒顯微系統架構 42
5.1主要光學元件特性與系統硬體架構 42
5.1.1 液晶相位延遲器基本特性 43
5.1.2 線性偏振片(Linear Polarizer)之穆勒矩陣與校正方式 44
5.1.3 偏振態產生器校正 46
5.2 穆勒矩陣量測方法 49
5.3 使用樣本偏振特性影像進行辨識的深度學習 52
第六章、 量測與訓練結果 57
6.1 肝細胞樣本 57
6.2 肝細胞樣本穆勒及分解 60
6.3 肝細胞偏振度影像深度學習結果 67
第七章、 結論與未來展望 70
第八章、 參考文獻 72

[1]“What Is Cancer? "National Cancer Institute, Sep. 17, 2007.
[2]H. Lodish, “Molecular Cell Biology,” New York, 2016.
[3]Stanley J. Swierzewski, “Hodgkin’s Disease - Staging,” Oct. 25, 2008.
[4]“Liver Cancer Stages.” Sep. 16, 2021.
[5]G. C. Giakos et al., “Infrared photon discrimination of lung cancer cells,” IEEE.IST, pp. 637-642, Jul. 2012, doi: 10.1109/IST.2012.6295599.
[6]S. Shrestha et al., “Label-free discrimination of lung cancer cells through mueller matrix decomposition of diffuse reflectance imaging,” Biomed. Signal Process. Control., 2018, doi: 10.1016/j.bspc.2017.05.009.
[7]V. Dremin et al., “Optical percutaneous needle biopsy of the liver: a pilot animal and clinical study,” Sci Rep, vol. 10, no. 1, p. 14200, Aug. 2020, doi: 10.1038/s41598-020-71089-5.
[8]J. Calderaro, M. Ziol, V. Paradis, and J. Zucman-Rossi, “Molecular and histological correlations in liver cancer,” J Hepatol, vol. 71, no. 3, pp. 616-630, Sep. 2019, doi: 10.1016/j.jhep.2019.06.001.
[9]L. Tchvialeva et al., “Polarization speckle imaging as a potential technique for in vivo skin cancer detection,” J Biomed Opt, vol. 18, no. 6, p. 061211, Dec. 2012, doi: 10.1117/1.JBO.18.6.061211.
[10]K. Banerjee and M. Mandal, “Oxidative stress triggered by naturally occurring flavone apigenin results in senescence and chemotherapeutic effect in human colorectal cancer cells,” Redox Biol., vol. 5, pp. 153-162, Aug. 2015, doi: 10.1016/j.redox.2015.04.009.
[11]J. Vizet et al., “In vivo imaging of uterine cervix with a Mueller polarimetric colposcope,” Sci Rep, vol. 7, no. 1, p. 2471, Jun. 2017, doi: 10.1038/s41598-017-02645-9.
[12]Y. Wang et al., “Mueller matrix microscope: a quantitative tool to facilitate detections and fibrosis scorings of liver cirrhosis and cancer tissues,” J Biomed Opt, vol. 21, no. 7, p. 71112, Jul. 2016, doi: 10.1117/1.JBO.21.7.071112.
[13]D. J. Evers et al., “Optical sensing for tumor detection in the liver,” Eur J Surg Oncol, vol. 39, no. 1, pp. 68-75, Jan. 2013, doi: 10.1016/j.ejso.2012.08.005.
[14]R. Pascanu, C. Gulcehre, K. Cho, and Y. Bengio, “How to Construct Deep Recurrent Neural Networks,” arXiv:1312.6026 , Apr. 2014,
[15]H. Mhaskar, Q. Liao, and T. Poggio, “When and Why Are Deep Networks Better Than Shallow Ones?,” AAAI, 17, pp. 2343-2349, Feb. 2017.
[16]S. Shrestha, J. Petermann, T. Farrahi, A. Deshpande, and G. C. Giakos, “Design, Calibration, and Testing of an Automated Near-Infrared Liquid-Crystal Polarimetric Imaging System for Discrimination of Lung Cancer Cells,” IEEE Transactions on Instrumentation and Measurement, vol. 64, no. 9, pp. 2453-2467, Sep. 2015, doi: 10.1109/TIM.2015.2415013.
[17]Y. Liu et al., “Detecting Cancer Metastases on Gigapixel Pathology Images,” arXiv:1703.02442 [cs], Mar. 2017, Accessed: Sep. 16, 2021.
[18]B. Liu, Y. Yao, R. Liu, H. Ma, and L. Ma, “Mueller polarimetric imaging for characterizing the collagen microstructures of breast cancer tissues in different genotypes,” Opt. Commun, vol. 433, pp. 60-67, Feb. 2019, doi: 10.1016/j.optcom.2018.09.037.
[19]L. Xia, Y. Yao, Y. Dong, M. Wang, H. Ma, and L. Ma, “Mueller polarimetric microscopic images analysis based classification of breast cancer cells,” Opt. Commun, vol. 475, p. 126194, Nov. 2020, doi: 10.1016/j.optcom.2020.126194.
[20]S.-Y. Lu and R. A. Chipman, “Interpretation of Mueller matrices based on polar decomposition,” J. Opt. Soc. Am, vol. 13, no. 5, pp. 1106-1113, May 1996, doi: 10.1364/JOSAA.13.001106.
[21]N. Ortega-Quijano and J. L. Arce-Diego, “Depolarizing differential Mueller matrices,” Opt. Lett, vol. 36, no. 13, pp. 2429-2431, Jul. 2011, doi: 10.1364/OL.36.002429.
[22]N. Ghosh, M. F. G. Wood, and I. A. Vitkin, “Mueller matrix decomposition for extraction of individual polarization parameters from complex turbid media exhibiting multiple scattering, optical activity, and linear birefringence,” J. Biomed. Opt, vol. 13, no. 4, p. 044036, Jul. 2008, doi: 10.1117/1.2960934.
[23]N. Ortega-Quijano and J. L. Arce-Diego, “Mueller matrix differential decomposition,” Opt. Lett, vol. 36, no. 10, p. 1942, May 2011, doi: 10.1364/OL.36.001942.
[24]R. M. A. Azzam, “Propagation of partially polarized light through anisotropic media with or without depolarization: A differential 4 × 4 matrix calculus,” J. Opt. Soc. Am, vol. 68, no. 12, pp. 1756-1767, Dec. 1978, doi: 10.1364/JOSA.68.001756.
[25]L. Deng and D. Yu, “Deep Learning: Methods and Applications,” Found. Trends Signal Process, Vol. 7, no. 3-4, pp. 197-387, May. 2014, doi:10.1561/2000000039
[26]M. Riesenhuber and T. Poggio, “Hierarchical models of object recognition in cortex,” Nat Neurosci, vol. 2, no. 11, pp. 1019-1025, Nov. 1999, doi: 10.1038/14819.
[27]D. Ciresan, U. Meier, J. Masci, and J. Schmidhuber, “Multi-Column Deep Neural Network for Traffic Sign Classification.” Neural Netw, vol. 32, pp. 333-338, Aug. 2012, doi: 10.1016/j.neunet.2012.02.023
[28]H. P. Martinez, Y. Bengio, and G. N. Yannakakis, “Learning deep physiological models of affect,” IEEE Comput. Intell. Mag., vol. 8, no. 2, pp. 20-33, May 2013, doi: 10.1109/MCI.2013.2247823.
[29]D. C. Ciresan, U. Meier, J. Masci, L. M. Gambardella, and J. Schmidhuber, “Flexible, High Performance Convolutional Neural Networks for Image Classification,” IJCAI,11, pp. 1237-1242, July 2011. doi: 10.5591/978-1-57735-516-8/IJCAI11-210


QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top