阿茲海默症的病症發展已經席捲全球逐漸影響全人們，而近年研究發現，不僅會導致記憶衰退，也可能會降低其他行為認知功能。現今大多數醫師使用量表和臨床數據來診斷病人是否患病。但是診斷的過程涉及到患者的配合度與醫師的主觀意識，因此對於病變的診斷產生偏差。本論文根據臨床數據和fMRI數據，將建置一個輔助醫師的自動辨識阿茲海默症系統，阿茲海默症數據來自於高雄醫學大學（KMU）的研究團隊。關於此系統的技術，主要分為三個階段：(1)數據前處理、(2)特徵擷取（相互信息、皮爾森相關係數、線性判別分析）、(3)分類模型（學習矢量量化、支持向量機），並通過實驗結果選擇最佳組合，成為阿茲海默症辨識系統。目前我們在臨床資料的分類上，已經達到不錯的準確度，我們還會探討資料做完共變量修正後的結果是否有助於診斷，並且透過特徵篩選的方法為醫生找出重要的特徵因子，藉此達到輔助醫師判斷阿茲海默症之目標。

Alzheimer’s disease has gradually affected human beings around the world. It not only causes the decay of memory, but may also decreases other cognitive functions. At present, most doctors use the and outpatient data to determine whether a patient has a disease. Because the judgment process involves the patient’s response and the doctor’s subjective decision, the resulting diagnosis can be questionable. This study intends to establish an automatic identification system which can help doctors to determine the Alzheimer’s disease based on the clinical and fMRI information of the patient from Kaohsiung Medical University (KMU). About the technology of the system, we will divide it into three-stage data pre-processing, feature extraction(Mutual Information, Pearson, Linear Discriminant Analysis), classification model(Learning Vector Quantization, Support Vector Machine), and select the best combination through the experimental results for Clinical Dementia Rating to become our system. We will also explore whether the results of Covariate correction for age, education affect the doctor’s judgment and provide doctors with more important features for patients.

論文審定書 i
誌謝 ii
摘　要 iii
Abstract iv
圖目錄 vii
表目錄 viii
第一章 導論 1
1.1. 研究背景與目的 1
1.2. 阿茲海默症介紹 2
1.3. 論文架構 5
第二章 文獻探討 7
2.1. 學習向量量化(Learning Vector Quantization, LVQ) 7
2.2. 支持向量機(Support Vector Machine, SVM) 8
第三章 研究方法 10
3.1. 利用MI、Pearson與LDA前處理 11
3.2. 將特徵篩選與合併的數據運用分類器進行分類比較其結果 13
3.3. 阿茲海默症患者的臨床資料進行前處理。 14
3.4. 因為數據的原始性，我們將運用KMEANS進行濾除雜訊 15
3.5. 分類出阿茲海默症病患與指標介紹 16
第四章 實驗結果 17
4.1. 特徵選取與縮減的實驗 19
4.2. 二階特徵選取與縮減的實驗 22
4.3. 臨床資料的實驗測試 23
4.4. 臨床資料濾除雜訊的實驗測試 23
4.5. 前處理臨床資料實驗測試以及維度排序 24
4.6. 與其他方法比較 25
第五章 結論與未來展望 26
5.1. 結論 26
5.2. 未來研究方向 26
參考文獻 27

[1] Proitsi, P., Kim, M., Whiley, L., Simmons, A., Sattlecker, M., Velayudhan, L., ... & Tsolaki, M. (2017). Association of blood lipids with Alzheimer's disease: A comprehensive lipidomics analysis. Alzheimer's & Dementia, 13(2), 140-151.
[2] Hussain, G., Anwar, H., Rasul, A., Imran, A., Qasim, M., Zafar, S., ... & Ahmad, W. (2019). Lipids as biomarkers of brain disorders. Critical reviews in food science and nutrition, 1-24.
[3] Bui, D. T., Tuan, T. A., Klempe, H., Pradhan, B., & Revhaug, I. (2016). Spatial prediction models for shallow landslide hazards: a comparative assessment of the efficacy of support vector machines, artificial neural networks, kernel logistic regression, and logistic model tree. Landslides, 13(2), 361-378.
[4] Pellegrini, E., Ballerini, L., Hernandez, M. D. C. V., Chappell, F. M., González-Castro, V., Anblagan, D., ... & Mair, G. (2018). Machine learning of neuroimaging for assisted diagnosis of cognitive impairment and dementia: A systematic review. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 10, 519-535.
[5] Ross, B. C. (2014). Mutual information between discrete and continuous data sets. PloS one, 9(2), e87357.
[6] Sacha, D., Zhang, L., Sedlmair, M., Lee, J. A., Peltonen, J., Weiskopf, D., ... & Keim, D. A. (2016). Visual interaction with dimensionality reduction: A structured literature analysis. IEEE transactions on visualization and computer graphics, 23(1), 241-250.
[7] Chen, T. T., & Lee, S. J. (2015). A weighted LS-SVM based learning system for time series forecasting. Information Sciences, 299, 99-116.
[8] Immanuel, M. S. H., & Jacob, S. G. (2017). Feature selection techniques for alzheimer’s disease: A review. International Journal of Engineering Technology Science and Research.
[9] Syahputra, R. (2016). APPLICATION OF NEURO-FUZZY METHOD FOR PREDICTION OF VEHICLE FUEL CONSUMPTION. Journal of Theoretical & Applied Information Technology, 86(1).
[10] Zvěřová, M. (2019). Clinical aspects of Alzheimer's disease. Clinical biochemistry.
[11] Alexiou, A., Kamal, M. A., & Ashraf, G. M. (2019). Alzheimer's Disease as a Current Challenge. Frontiers in neuroscience, 13, 768.
[12] Chang, Y. L., Chen, T. F., Shih, Y. C., Chiu, M. J., Yan, S. H., & Tseng, W. Y. I. (2015). Regional cingulum disruption, not gray matter atrophy, detects cognitive changes in amnestic mild cognitive impairment subtypes. Journal of Alzheimer's Disease, 44(1), 125-138.
[13] Hammer, B., & Villmann, T. (2002). Generalized relevance learning vector quantization. Neural Networks, 15(8-9), 1059-1068.
[14] Kohonen, T. (1998). Learning vector quantization, The handbook of brain theory and neural networks.
[15] Schneider, P., Biehl, M., & Hammer, B. (2009). Adaptive relevance matrices in learning vector quantization. Neural computation, 21(12), 3532-3561.
[16] Smola, A. J., & Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and computing, 14(3), 199-222.
[17] W. G. Cobourn, “Accuracy and reliability of an automated air quality forecast system for ozone in seven Kentucky metropolitan areas,” Atmospheric Environment, vol. 41, no. 28, pp. 5863–5875, 2007.
[18] Haykin, S. S. (2009). Neural networks and learning machines/Simon Haykin. New York: Prentice Hall,.
[19] Viola, P., & Wells III, W. M. (1997). Alignment by maximization of mutual information. International journal of computer vision, 24(2), 137-154.
[20] Kraskov, A., Stögbauer, H., & Grassberger, P. (2004). Estimating mutual information. Physical review E, 69(6), 066138.
[21] Battiti, R. (1994). Using mutual information for selecting features in supervised neural net learning. IEEE Transactions on neural networks, 5(4), 537-550.
[22] Lee Rodgers, J., & Nicewander, W. A. (1988). Thirteen ways to look at the correlation coefficient. The American Statistician, 42(1), 59-66.
[23] Benesty, J., Chen, J., Huang, Y., & Cohen, I. (2009). Pearson correlation coefficient. In Noise reduction in speech processing (pp. 1-4). Springer, Berlin, Heidelberg.
[24] Saad, Z. S., Glen, D. R., Chen, G., Beauchamp, M. S., Desai, R., & Cox, R. W. (2009). A new method for improving functional-to-structural MRI alignment using local Pearson correlation. Neuroimage, 44(3), 839-848.
[25] Prince, S. J., & Elder, J. H. (2007, October). Probabilistic linear discriminant analysis for inferences about identity. In 2007 IEEE 11th International Conference on Computer Vision (pp. 1-8). IEEE.
[26] Izenman, A. J. (2013). Multivariate regression. In Modern Multivariate Statistical Techniques (pp. 159-194). Springer, New York, NY.
[27] Ye, J., Janardan, R., & Li, Q. (2005). Two-dimensional linear discriminant analysis. In Advances in neural information processing systems (pp. 1569-1576).
[28] Zhu, X., Zhang, S., Jin, Z., Zhang, Z., & Xu, Z. (2010). Missing value estimation for mixed-attribute data sets. IEEE Transactions on Knowledge and Data Engineering, 23(1), 110-121.
[29] Little, R. J. (1988). Missing-data adjustments in large surveys. Journal of Business & Economic Statistics, 6(3), 287-296.
[30] Gerard, H. B., Wilhelmy, R. A., & Conolley, E. S. (1968). Conformity and group size. Journal of Personality and Social Psychology, 8(1p1), 79.
[31] Baron, R. S., Vandello, J. A., & Brunsman, B. (1996). The forgotten variable in conformity research: Impact of task importance on social influence. Journal of personality and social psychology, 71(5), 915.
[32] Pennanen, C., Kivipelto, M., Tuomainen, S., Hartikainen, P., Hänninen, T., Laakso, M. P., ... & Vainio, P. (2004). Hippocampus and entorhinal cortex in mild cognitive impairment and early AD. Neurobiology of aging, 25(3), 303-310.
[33] Fratiglioni, L., Grut, M., Forsell, Y., Viitanen, M., Grafström, M., Holmen, K., ... & Winblad, B. (1991). Prevalence of Alzheimer's disease and other dementias in an elderly urban population: relationship with age, sex, and education. Neurology, 41(12), 1886-1886.
[34] Breitner, J. C., Wyse, B. W., Anthony, J. C., Welsh-Bohmer, K. A., Steffens, D. C., Norton, M. C., ... & Khachaturian, A. (1999). APOE-ε4 count predicts age when prevalence of AD increases, then declines: the Cache County Study. Neurology, 53(2), 321-321.
[35] Wolfensberger, W. P., Nirje, B., Olshansky, S., Perske, R., & Roos, P. (1972). The principle of normalization in human services.
[36] Han, J., Pei, J., & Kamber, M. (2011). Data mining: concepts and techniques. Elsevier.
[37] Bramer, M. (2007). Principles of data mining (Vol. 180). London: Springer.
[38] Jain, A. K. (2010). Data clustering: 50 years beyond K-means. Pattern recognition letters, 31(8), 651-666..
[39] Wagstaff, K., Cardie, C., Rogers, S., & Schrödl, S. (2001, June). Constrained k-means clustering with background knowledge. In Icml (Vol. 1, pp. 577-584).
[40] Alsabti, K., Ranka, S., & Singh, V. (1997). An efficient k-means clustering algorithm.
[41] Townsend, J. T. (1971). Theoretical analysis of an alphabetic confusion matrix. Perception & Psychophysics, 9(1), 40-50.
[42] Asuncion, A., & Newman, D. (2007). UCI machine learning repository.
[43] https://archive.ics.uci.edu/ml/index.php
[44] Zhou, Q., Goryawala, M., Cabrerizo, M., Wang, J., Barker, W., Loewenstein, D. A., ... & Adjouadi, M. (2014). An optimal decisional space for the classification of Alzheimer's disease and mild cognitive impairment. IEEE Transactions on Biomedical Engineering, 61(8), 2245-2253.
[45] Rieger, S. A., Muraleedharan, R., & Ramachandran, R. P. (2014, September). Speech based emotion recognition using spectral feature extraction and an ensemble of kNN classifiers. In The 9th International Symposium on Chinese Spoken Language Processing (pp. 589-593). IEEE.
[46] Kamiński, B., Jakubczyk, M., & Szufel, P. (2018). A framework for sensitivity analysis of decision trees. Central European journal of operations research, 26(1), 135-159.
[47] Tang, J., Deng, C., & Huang, G. B. (2015). Extreme learning machine for multilayer perceptron. IEEE transactions on neural networks and learning systems, 27(4), 809-821.