結合臨床資料與舌下脈絡影像的機器學習技術應 用於肝病種類之判讀

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：45

、訪客IP：3.16.51.221

姓名

黃映瑄(Ying-Shiuan Huang) 查詢紙本館藏

畢業系所

數學系

論文名稱

結合臨床資料與舌下脈絡影像的機器學習技術應用於肝病種類之判讀
(Classification of Liver Diseases Through Machine Learning Techniques Combining Clinical Data and Sublingual Vein Imaging)

相關論文

★ 非線性塊狀高斯消去牛頓演算法在噴嘴流體的應用	★ 以平行 Newton-Krylov-Schwarz 演算法解 Poisson-Boltzmann 方程式的有限元素解在膠體科學上的應用
★ 最小平方有限元素法求解對流擴散方程以及使用Bubble函數的改良	★ Bifurcation Analysis of Incompressible Sudden Expansion Flows Using Parallel Computing
★ Parallel Jacobi-Davidson Algorithms and Software Developments for Polynomial Eigenvalue Problems in Quantum Dot Simulation	★ An Inexact Newton Method for Drift-DiffusionModel in Semiconductor Device Simulations
★ Numerical Simulation of Three-dimensional Blood Flows in Arteries Using Domain Decomposition Based Scientific Software Packages in Parallel Computers	★ A Parallel Fully Coupled Implicit Domain Decomposition Method for the Stabilized Finite Element Solution of Three-dimensional Unsteady Incompressible Navier-Stokes Equations
★ A Study for Linear Stability Analysis of Incompressible Flows on Parallel Computers	★ Parallel Computation of Acoustic Eigenvalue Problems Using a Polynomial Jacobi-Davidson Method
★ Numerical Study of Algebraic Multigrid Methods for Solving Linear/Nonlinear Elliptic Problems on Sequential and Parallel Computers	★ A Parallel Multilevel Semi-implicit Scheme of Fluid Modeling for Numerical Low-Temperature Plasma Simulation
★ Performance Comparison of Two PETSc-based Eigensolvers for Quadratic PDE Problems	★ A Parallel Two-level Polynomial Jacobi-Davidson Algorithm for Large Sparse Dissipative Acoustic Eigenvalue Problems
★ A Full Space Lagrange-Newton-Krylov Algorithm for Minimum Time Trajectory Optimization	★ Parallel Two-level Patient-specific Numerical Simulation of Three-dimensional Rheological Blood Flows in Branching Arteries

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

本研究基於機器學習運用臨床資料與舌下靜脈影像，發展電腦輔助診斷(Computer Aided Diagnosis, CAD)協助醫師判讀肝病種類。慢性肝病包含慢性肝炎、肝硬化以及肝癌，而亞洲為肝病的好發區域。實驗考慮三個二元分類問題，分別區分肝癌、肝硬化、脂肪肝與健康者。病患的臨床數據包含基本資料、抽血檢驗值、是否有抽菸、飲酒、喝咖啡的習慣；中醫相關研究指出舌下靜脈曲張與人體疾病存在關聯性，肝硬化患者的舌下靜脈曲張程度較健康者更嚴重，因此本研究除了利用臨床數據，也加入患者的舌部影像作為訓練特徵。我們使用主成分分析(Principal Component Analysis, PCA)與卷積神經網路(Convolutional Neural Network, CNN)對舌下影像提取特徵，並利用特徵選取(Feature Selection)對臨床數據篩選重要特徵，最後結合這些重要特徵作為訓練機器學習模型之用。我們利用了Random Forest、Support Vector Machines、K-Nearest Neighbors、Ridge、Logistic Regression、Multilayer Perceptron等六種機器學習方法進行實驗。實驗結果顯示，僅使用臨床資料時，機器學習對區分健康與脂肪肝者(脂肪肝：80.7%)相對於其他兩個二元分類問題(肝癌：74%,肝硬化：64.5%)有較高的準確率；另外，僅使用舌下脈絡影像當作特徵時，機器學習模型對於肝癌與肝硬化在統計有顯著差異(肝癌：52.4%、肝硬化：55.1%)，且準確率明顯低於僅使用臨床資料。最後，利用影像結合臨床資料，經特徵提取、特徵選取與網格搜尋分別在肝癌、肝硬化、脂肪肝得到準確率77.8%、70.52%、82.6%，相較於僅用臨床資料為特徵時，模型準確率提升2~6%。其中，檢測脂肪肝時在Random Forest模型下，運用CNN與反向特徵消除法(Backward Feature Elimination)，最佳準確率達85.6%；檢測肝硬化時在Multilayer Perceptron模型下，運用CNN與Ridge特徵選取法，最佳準確率達73.4%；檢測肝癌時在Multilayer Perceptron模型下，運用CNN與反向特徵消除法，最佳準確率達81.8%。現代醫學仰賴血液檢查與腹部超音波偵測肝臟疾病，我們提出的機器學習方法可以作為醫學中診斷是否有肝臟疾病的第二意見，克服僅用血液檢驗的診斷侷限。

摘要(英)

Chronic liver disease includes chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma. Also, liver disease is a severe disease, especially in Asia. In this study, we consider three binary classifications which are hepatocellular carcinoma(HCC), liver cirrhosis(LC), and fatty liver disease(FLD). All of these patients are compared with people without these diseases. There are two types of data in our data set. One is clinical or laboratory numerical data, and the other is sublingual vein imaging. We use feature selection techniques to select salient features from numerical data. Furthermore, we utilize principal component analysis (PCA) and convolutional neural network (CNN) for feature extraction from imaging. Following this, we combine these two sources of information and employ six machine learning algorithms, including Random Forest(RF), Support Vector Machines(SVM), K-Nearest Neighbors(KNN), Ridge, Logistic Regression(LR), and Multilayer Perceptron(MLP) for training.The results show that there is higher accuracy for classifying FLD(80.7%) than HCC(74.0%) and LC(64.5%) by utilizing numerical data. In addition, imaging has accuracy for classifying HCC and LC, with results as statistically significant(HCC: 52.4%, LC: 55.1%). In the case of HCC and LC, the accuracy is lower by utilizing imaging than by utilizing numerical data. Finally, we combine two data types and employ feature extraction, feature selection, and grid search in our model. The mean accuracy for HCC, LC, and FLD is 77.8%, 70.52%, and 82.6%, respectively. We improve the mean of the accuracy by 2~6% by combining features. Moreover, the best accuracy is 81.8% for HCC through MLP, CNN, and backward elimination; the best accuracy is 73.4% for LC through MLP, CNN, and RidgeCV; the best accuracy is 85.6% for FLD through RF, CNN, and backward elimination. In conclusion, in addition to blood tests and ultrasound examinations to detect liver disease, the proposed method can be potentially used as a second opinion to overcome the limitations of classical diagnostic approaches.

關鍵字(中)

★ 機器學習
★ 卷積神經網路
★ 二元分類
★ 臨床資料
★ 舌下靜脈影像

關鍵字(英)

★ Machine Learning
★ Convolutional Neural Networks
★ Binary Classification
★ Clinical Data
★ Sublingual Vein Imaging

論文目次

致謝 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
1 緒論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 方法 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1 機器學習 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1.1 脊迴歸 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1.2 隨機森林法 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.3 K 近鄰演算法 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.4 支持向量機 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.5 多層感知器 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.6 邏輯迴歸法 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 卷積神經網路 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2.1 卷積層 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2.2 池化層 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.3 全連接層 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 實驗設計 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.1 資料來源 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.2 預處理 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.1 影像切割與正規化 . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.2 缺失資料插補處理 . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.3 特徵縮放 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3 特徵工程 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.1 維度降低 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.2 特徵選取 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.4 交叉驗證 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4 結果與討論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.1 僅影像特徵 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2 僅數值特徵 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.3 混成不同型態的特徵 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5 結論與未來相關研究建議 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
附錄 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

參考文獻

[1] J. Ferlay, M. Colombet, I. Soerjomataram, D. M. Parkin, M. Piñeros, A. Znaor,
and F. Bray. Cancer statistics for the year 2020: An overview. Int. J. Cancer,
149:778–789, 2021.
[2] J. D. Yang, P. Hainaut, G. J. Gores, A. Amadou, A. Plymoth, and L. R. Roberts. A
global view of hepatocellular carcinoma: trends, risk, prevention and management.
Nat. Rev. Gastroenterol. Hepatol., 16:589–604, 2019.
[3] E. S. Bialecki, A. M. Ezenekwe, E. M. Brunt, B. T. Collins, T. B. Ponder, B. K.
Bieneman, and A. M. Di Bisceglie. Comparison of liver biopsy and noninvasive
methods for diagnosis of hepatocellular carcinoma. Clin. Gastroenterol. Hepatol.,
4:361–368, 2006.
[4] L. A. Adams and Keith D. Angulo, P. L. Nonalcoholic fatty liver disease. Can. Med.
Assoc. J., 172:899–905, 2005.
[5] O. J. Kennedy, J. A. Fallowfield, R. Poole, P. C. Hayes, J. Parkes, and P. J. Roderick.
All coffee types decrease the risk of adverse clinical outcomes in chronic liver disease:
A UK biobank study. BMC Public Health, 21:1–14, 2021.
[6] A. Vallet-Pichard, V. Mallet, B. Nalpas, V. Verkarre, A. Nalpas, V. Dhalluin-Venier,
H. Fontaine, and S. Pol. FIB-4: an inexpensive and accurate marker of fibrosis in
HCV infection. comparison with liver biopsy and fibrotest. Hepatology, 46:32–36,
2007.
[7] A. Spann, A. Yasodhara, J. Kang, K. Watt, B. Wang, A. Goldenberg, and M. Bhat.
Applying machine learning in liver disease and transplantation: a comprehensive
review. Hepatology, 71:1093–1105, 2020.
[8] L.-Q. Zhou, J.-Y. Wang, S.-Y. Yu, G.-G. Wu, Q. Wei, Y.-B. Deng, X.-L. Wu, X.-W.
Cui, and C. F. Dietrich. Artificial intelligence in medical imaging of the liver. World
J. Gastroenterol., 25:672, 2019.
[9] A. Hosny, C. Parmar, J. Quackenbush, L. H. Schwartz, and H. J. Aerts. Artificial
intelligence in radiology. Nat. Rev. Cancer, 18:500–510, 2018.
33
[10] T. F. Chong, K. T. Wong, and Y. P. Yong. The status and prospective study of
tongue diagnostic in traditional Chinese medicine. Inti J., 2022:1–7, 2022.
[11] B. Kirschbaum. Atlas of Chinese Tongue Diagnosis, volume 1. Eastland Press, 2000.
[12] J. Zhao, L. Y. Guo, J. M. Yang, and J. W. Jia. Sublingual vein parameters, AFP,
AFP-L3, and GP73 in patients with hepatocellular carcinoma. Genet. Mol. Res.,
14:7062–7067, 2015.
[13] M. Tandon, H. Singh, N. Singla, P. Jain, and C. K. Pandey. Tongue thickness in
health vs cirrhosis of the liver: Prospective observational study. World J. Gastrointest. Pharmacol. Ther., 11:59, 2020.
[14] Q. Liu, X.-Q. Yue, R.-Z. Ren, C.-H. Ma, and C.-Q. Ling. Characteristics of sublingual
venae in primary liver cancer patients in different clinical stages. Chin. J. Integr.
Med., 2:175–177, 2004.
[15] L.-C. Lo, C. Chen, J. Y. Chiang, T.-L. Cheng, H.-J. Lin, and H.-H. Chang. Tongue
diagnosis of traditional Chinese medicine for rheumatoid arthritis. Afr. J. Tradit.
Complement. Altern. Med., 10:360–369, 2013.
[16] N. Sandhya and M. Rajasekar. Tongue image analysis for hepatitis detection using
SVM. Indian J. Comput. Sci. Eng., 8:526–534, 2017.
[17] P.-C. Hsu, H.-K. Wu, Y.-C. Huang, H.-H. Chang, T.-C. Lee, Y.-P. Chen, J. Y.
Chiang, and L.-C. Lo. The tongue features associated with type 2 diabetes mellitus.
Med., 98, 2019.
[18] L.-C. Lo, T.-L. Cheng, Y.-J. Chen, S. Natsagdorj, and J. Y. Chiang. TCM tongue
diagnosis index of early-stage breast cancer. Complement. Ther. Med., 23:705–713,
2015.
[19] X. Xu, H.-L. Zhang, Q.-P. Liu, S.-W. Sun, J. Zhang, F.-P. Zhu, G. Yang, X. Yan,
Y.-D. Zhang, and X.-S. Liu. Radiomic analysis of contrast-enhanced CT predicts microvascular invasion and outcome in hepatocellular carcinoma. J. Hepatol., 70:1133–
1144, 2019.
[20] S. Nowak, N. Mesropyan, A. Faron, W. Block, M. Reuter, U. I. Attenberger, J. A.
Luetkens, and A. M. Sprinkart. Detection of liver cirrhosis in standard T2-weighted
MRI using deep transfer learning. Eur. Radiol., 31:8807–8815, 2021.
34
[21] K. Yasaka, H. Akai, A. Kunimatsu, O. Abe, and S. Kiryu. Liver fibrosis: deep convolutional neural network for staging by using gadoxetic acid–enhanced hepatobiliary
phase MR images. Radiology, 287:146–155, 2018.
[22] I. Gatos, S. Tsantis, S. Spiliopoulos, D. Karnabatidis, I. Theotokas, P. Zoumpoulis,
T. Loupas, J. D. Hazle, and G. C. Kagadis. A machine-learning algorithm toward
color analysis for chronic liver disease classification, employing ultrasound shear wave
elastography. Ultrasound Med. Biol., 43:1797–1810, 2017.
[23] M. Byra, G. Styczynski, C. Szmigielski, P. Kalinowski, Ł. Michałowski,
R. Paluszkiewicz, B. Ziarkiewicz-Wróblewska, K. Zieniewicz, P. Sobieraj, and
A. Nowicki. Transfer learning with deep convolutional neural network for liver
steatosis assessment in ultrasound images. Int. J. Comput. Assist. Radiol. Surg.,
13:1895–1903, 2018.
[24] C.-C. Wu, W.-C. Yeh, W.-D. Hsu, M. M. Islam, P. A. A. Nguyen, T. N. Poly, Y.-C.
Wang, H.-C. Yang, and Y.-C. J. Li. Prediction of fatty liver disease using machine
learning algorithms. Comput. Methods Programs Biomed., 170:23–29, 2019.
[25] M. Islam, C.-C. Wu, T. N. Poly, H.-C. Yang, Y.-C. J. Li, et al. Applications of
machine learning in fatty live disease prediction. In Building Continents of Knowledge
in Oceans of Data: The Future of Co-Created eHealth, pages 166–170. IOS Press,
2018.
[26] T.-F. Yip, A. Ma, V.-S. Wong, Y.-K. Tse, H.-Y. Chan, P.-C. Yuen, and G.-H. Wong.
Laboratory parameter-based machine learning model for excluding non-alcoholic
fatty liver disease (NAFLD) in the general population. Aliment. Pharmacol. Ther.,
46:447–456, 2017.
[27] Y. Cao, Z.-D. Hu, X.-F. Liu, A.-M. Deng, and C.-J. Hu. An MLP classifier for prediction of HBV-induced liver cirrhosis using routinely available clinical parameters.
Dis. Markers, 35:653–660, 2013.
[28] Y. Cao, K. He, M. Cheng, H.-Y. Si, H.-L. Zhang, W. Song, A.-L. Li, C.-J. Hu, and
N. Wang. Two classifiers based on serum peptide pattern for prediction of HBVinduced liver cirrhosis using MALDI-TOF MS. Biomed Res. Int., 2013, 2013.
35
[29] Y. Wang, L. Ma, and P. Liu. Feature selection and syndrome prediction for liver
cirrhosis in traditional Chinese medicine. Comput Methods Programs Biomed, 95:249–
257, 2009.
[30] D. Song, Y. Wang, W. Wang, Y. Wang, J. Cai, K. Zhu, M. Lv, Q. Gao, J. Zhou,
J. Fan, et al. Using deep learning to predict microvascular invasion in hepatocellular carcinoma based on dynamic contrast-enhanced MRI combined with clinical
parameters. J. Cancer Res. Clin. Oncol., 147:3757–3767, 2021.
[31] A. Géron. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow:
Concepts, Tools, and Techniques to Build Intelligent Systems. O’Reilly Media, 2019

指導教授

黃楓南(Feng-Nan Hwang)

審核日期

2022-9-3

推文