根據世界衛生組織的統計，憂鬱症是全球失能的首要原因，其中重度憂慮症的患者通常會共病一種以上的焦慮症或其他精神疾病。憂鬱症的臨床診斷非常依賴醫師進行診斷式會談及患者填寫的醫學量表作為依據，為了防止人為因素影響，所以需要客觀與量化的診斷工具解決此問題。因此本論文的目的為利用腦電訊號發展一套分辨重度憂鬱共病焦慮症的患者與健康對照組的機器學習方法。
為了表示腦電訊號（electroencephalogram, EEG）的特性，首先計算EEG各頻帶之相對能量，再透過主成份分析（principal component analysis, PCA）的方法，開發多項EEG訊號複雜度的特徵，以及利用圖論分析來量化大腦的功能連結程度，並得到各頻道間互動性的特徵變數。建立模型方面，首先透過特徵篩選工程的技術篩選特徵後，再將剩下的特徵建立區分患者與健康人的極限梯度提升（extreme gradient boosting, XGBoost）機器學習分類模型。最後以2項睡眠量表的分數作為特徵，結果中顯示與EEG的特徵合併訓練後，模型的分類效能有顯著的提升。
本文之實驗對象為135位重度憂鬱共病焦慮症的患者與135位健康對照組，資料集透過EEG特徵與圖論建立的模型，分類精度為0.827、敏感度為0.803、特異度為0.852。單以睡眠量表的2種問卷分數為特徵時，分類精度為0.927、敏感度為0.915、特異度為0.935。使用EEG特徵配合睡眠量表訓練的精度則可達0.952、敏感度為0.947、特異度為0.957。
頻帶能量的分析結果顯示患者與健康人的大腦活動在額葉區的差異相較於其它區域更為明顯。PCA的結果中發現患者的EEG活動模式相較健康人更為複雜。圖論分析的結果顯示，患者大腦網路的整合性較差，相較於健康人缺乏小世界網路的特性。

According to World Health Organization, depression is the global leading cause of disability. Major depressive disorder (MDD) is often comorbid with anxiety disorders or symptoms. Conventionally, the diagnosis of such illnesses relies mainly on clinical interview and psychiatric questionnaires. To tackle this problem more objectively and quantitatively, the goal of this study is to develop machine learning methods to differentiate patients comorbid with major depressive disorder and anxiety symptoms from healthy controls by using electroencephalogram (EEG) signals.
To characterize the EEG signals, we first computed the relative energies of different frequency bands. Based on these results, we used the principal component analysis (PCA) method to develop a number of features to characterize the complexity of the EEG signals. Additional features were developed by using graph theory to quantify functional connectivity between different channels of EEG signals. After filtering these features with feature engineering techniques, the extreme gradient boosting (XGBoost) machine learning classification model was used to differentiate patients and healthy controls. Finally, we showed the classification results can be considerably improved by appending two sleep quality questionnaire scores into the feature set.
The test subjects include 135 patients and 135 healthy controls. Classifying these two groups of subjects with only the EEG features, the resulting accuracy, sensitivity and specificity were 0.827, 0.803 and 0.852, respectively. With two sleep quality questionnaire scores as the features, the accuracy, sensitivity and specificity were 0.927, 0.915 and 0.935, respectively. By using both EEG features and sleep quality questionnaire scores, the accuracy, sensitivity and specificity improved to 0.952, 0.947 and 0.957, respectively.
Qualitatively, based on frequency band power analysis, our results show that the difference of the frontal lobe region activities between the patients and controls is more pronounced than that of other brain regions. The results of the PCA analysis demonstrate that patients had a more complex EEG activity pattern than healthy controls. The graph theory results demonstrated that patients showed a loss of small-world network characteristics.

目錄
論文審定書 i
誌謝 ii
摘要 iv
Abstract v
目錄 vii
圖目錄 xii
表目錄 xv
第一章 緒論 1
1.1 前言 1
1.2 文獻回顧 1
1.2.1腦波分析 1
1.2.2圖形理論網路分析 2
1.2.3機器學習系統 4
1.3 研究目的與方法 7
1.4 論文架構 7
第二章 實驗方法 9
2.1 研究方法 9
2.2 實驗量表 11
2.3 腦波測量儀 13
2.4 大腦皮質與腦電圖簡介 14
2.4.1 大腦皮質區域 14
2.4.2 腦電圖（Electroencephalography, EEG） 16
2.5 訊號前處理 17
2.6 實驗程序 18
第三章 腦波訊號分析方法 20
3.1 相關係數矩陣（Correlation coefficient matrix） 20
3.2 頻帶能量分析 21
3.2.1 Bartlett's method[70] 22
3.2.2 Welch's method[71] 22
3.2.3 快速傅立葉轉換（Fast Fourier transform）[72] 23
3.2.4 墊零法（Zero padding） 25
3.3 主成份分析（Principal component analysis）[73] 25
3.4 訊號複雜度分析方法 30
3.4.1 相關結構（Correlation structure） 30
3.4.2 熵（Entropy）[79] 32
3.5 圖形理論（Graph theory）網路分析 33
3.5.1 功能連通性（Functional connectivity） 33
3.5.2 Dijkstra最短路徑演算法[87] 36
3.5.3 網路分析方法 38
3.5.4 小世界網路（Small-world networks） 42
第四章 機器學習特徵與模型 44
4.1 時域訊號特徵 45
4.1.1相關係數矩陣 45
4.1.2 相關矩陣特徵值 46
4.2 頻帶能量特徵 47
4.2.1 絕對能量 48
4.2.2 相對能量 49
4.2.3 頻道之間能量比例同步性 49
4.3 主成份分析特徵 50
4.3.1主軸投影量 51
4.3.2 特徵值（Eigenvalue） 51
4.3.3 主軸特徵向量（Eigenvector）相似度 52
4.3.4 頻道之間投影量變化同步性 53
4.4 圖形理論特徵 54
4.5 機器學習模型 56
4.5.1 機器學習簡介[94] 56
4.5.2 集成學習（Ensemble learning） 57
4.5.3 極限梯度提升（Extreme Gradient Boosting, XGBoost）[97] 59
4.5.4 特徵篩選方法 63
4.5.5 交叉驗證（K-fold Cross-validation）參數調整方法 [100] 65
4.5.6 模型訓練流程 68
4.6 效能評估方法 71
4.6.1 分類效能指標 71
4.6.2 個案評估方法 72
4.7 研究使用之硬體規格與軟體 73
4.7.1 硬體規格 73
4.7.2 軟體介紹 73
第五章 分析結果與討論 74
5.1 分析項目規劃 74
5.2 腦波分析結果 75
5.2.1 時域訊號分析結果 75
5.2.2 頻帶能量分析結果 79
5.2.3 主成份分析之結果 88
5.2.4 圖形理論網路分析結果 94
5.3 機器學習分類結果 97
5.3.1 腦波特徵分類效能 98
5.3.2 實驗階段各期之分類效能 101
5.3.3 腦波特徵與圖形理論特徵分類效能 104
5.3.4 基準期輸出值與回想期輸出值分類效能 105
5.3.5 基準期輸出值與睡眠量表分類效能 106
第六章 結論與未來展望 109
6.1 結論 109
6.2 未來展望 110
參考文獻 111
附錄I 相對能量分析結果 123
附錄II 腦波頻帶PCA分析結果 129
附錄III PCA主軸相似度結果 137
附錄IV 圖論分析結果 143

參考文獻
1. Jiang, X., Shen, Y., Yao, J., Zhang, L., Xu, L., Feng, R., Cai, L., Liu, J., Chen, W., & Wang, J. (2019). Connectome analysis of functional and structural hemispheric brain networks in major depressive disorder. Translational Psychiatry, 9(1), 136.
2. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders. Arlington, VA: American Psychiatric Publishing.
3. Ionescu, D. F., Niciu, M. J., Mathews, D. C., Richards, E. M., & Zarate Jr, C. A. (2013). Neurobiology of anxious depression: A review. Depression and Anxiety, 30(4), 374-385.
4. Davidson, R. J. (1998). Anterior electrophysiological asymmetries, emotion, and depression: Conceptual and methodological conundrums. Psychophysiology, 35(5), 607-614.
5. Davidson, R. J. (1992). Anterior cerebral asymmetry and the nature of emotion. Brain and Cognition, 20(1), 125-151.
6. Baehr, E., Rosenfeld, J. P., Baehr, R., & Earnest, C. (1998). Comparison of two EEG asymmetry indices in depressed patients vs. normal controls. International Journal of Psychophysiology, 31(1), 89-92.
7. Briesemeister, B. B., Tamm, S., Heine, A., & Jacobs, A. M. (2013). Approach the good, withdraw from the bad-A Review on frontal alpha asymmetry measures in applied psychological research. Psychology, 4(3), 261-267.
8. Gotlib, I. H., Ranganathand, C., & Rosenfeld, J. P. (1998). Frontal EEG alpha asymmetry,depression,and cognitive functioning. Cognition and Emotion, 12(3), 449-478.
9. Jesulola, E., Sharpley, C. F., Bitsika, V., Agnew, L. L., & Wilson, P. (2015). Frontal alpha asymmetry as a pathway to behavioural withdrawal in depression: Research findings and issues. Behavioural Brain Research, 292, 56-67.
10. Kentgen, L. M., Tenke, C. E., Pine, D. S., Fong, R., Klein, R. G., & Bruder, G. E. (2000). Electroencephalographic asymmetries in adolescents with major depression: Influence of comorbidity with anxiety disorders. Journal of Abnormal Psychology, 109(4), 797.
11. Nitschke, J. B., Heller, W., Palmieri, P. A., & Miller, G. A. (1999). Contrasting patterns of brain activity in anxious apprehension and anxious arousal. Psychophysiology, 36(5), 628-637.
12. Heller, W., Nitschke, J. B., Etienne, M. A., & Miller, G. A. (1997). Patterns of regional brain activity differentiate types of anxiety. Journal of Abnormal Psychology, 106(3), 376.
13. Davidson, R. J., Chapman, J. P., & Chapman, L. J. (1987). Task-dependent EEG asymmetry discriminates between depressed and non-depressed subjects. Psychophysiology, 24(1), 585.
14. Henriques, J. B., & Davidson, R. J. (1990). Regional brain electrical asymmetries discriminate between previously depressed and healthy control subjects. Journal of Abnormal Psychology, 99, 22-31.
15. Reid, S. A., Duke, L. M., & Allen, J. J. (1998). Resting frontal electroencephalographic asymmetry in depression: Inconsistencies suggest the need to identify mediating factors. Psychophysiology, 35(4), 389-404.
16. Debener, S., Beauducel, A., Nessler, D., Brocke, B., Heilemann, H., & kayser, J. (2000). Is resting anterior EEG alpha asymmetry a trait marker for depression? Neuropsychobiology, 41(1), 31-37.
17. Nitschke, J. B., Heller, W., Palmieri, P. A., & Miller, G. A. (1999). Contrasting patterns of brain activity in anxious apprehension and anxious arousal. Psychophysiology, 36(5), 628-637.
18. Yamada, M., Kimura, M., Mori, T., & Endo, S. (1995). EEG power and coherence in presenile and senile depression. Characteristic findings related to differences between anxiety type and retardation type. Nihon Ika Daigaku Zasshi, 62(2), 176-185.
19. Pizzagalli, D. A., Nitschke, J. B., Oakes, T. R., Hendrick, A. M., Horras, K. A., Larson, C. L., & Benca, R. M. (2002). Brain electrical tomography in depression: The importance of symptom severity, anxiety, and melancholic features. Biological Psychiatry, 52(2), 73-85.
20. Fingelkurts, A. A., Fingelkurts, A. A., Rytsälä, H., Suominen, K., Isometsä, E., & Kähkönen, S. (2006). Composition of brain oscillations in ongoing EEG during major depression disorder. Neuroscience Research, 56(2), 133-144.
21. Coan, J. A., Allen, J. J., & McKnight, P. E. (2006). A capability model of individual differences in frontal EEG asymmetry. Biological Psychology, 72(2), 198-207.
22. Lin, I. M., Lin, H. Y., Wang, S. Y., Fan, S. Y., Lee, Y., Chiu, N. M., Hung, C. F., Yen, C. F., Yeh, Y. C., Huang, M. F., Liu, T. L., Wang, P. W., & Lin, H. C. (2019). Quantitative electroencephalography in patients with major depressive disorder combined with anxious symptoms. Poster session presented at the Annual Meeting of Biofeedback Foundation of Europe, Cardiff, Wales, UK.
23. 林欣儀（2018）。共病焦慮症狀的重度憂鬱症患者之量化腦波研究。高雄醫學大學碩士論文，高雄市。
24. Lin, I. M., Chen, T. C., Lin, H. Y., Wang, S. Y., Sung, J. L., & Yen, C. W. (2021). Electroencephalogram patterns in patients comorbid with major depressive disorder and anxiety symptoms: Proposing a hypothesis based on hypercortical arousal and not frontal or parietal alpha asymmetry. Journal of affective disorders, 282, 945-952.
25. Watts, D. J., & Strogatz, S. H. (1998). Collective dynamics of 'small-world' networks. Nature, 393(6684), 440-442.
26. Suo, X., Lei, D., Li, L., Li, W., Dai, J., Wang, S., He, M., Zhu, H., Kemp, G.J., & Gong, Q. (2018). Psychoradiological patterns of small-world properties and a systematic review of connectome studies of patients with 6 major psychiatric disorders. Journal of Psychiatry & Neuroscience, 43(6), 416-427.
27. Sun, Y., Hu, S., Chambers, J., Zhu, Y., & Tong, S. (2011). Graphic patterns of cortical functional connectivity of depressed patients on the basis of EEG measurements. Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1419-1422.
28. Kalpana, R., & Gnanambal, I. (2016). The analysis of nonlinear invariants of multi-channel EEG signal using graph-theory connectivity approach in patient with depression. Asian Journal of Information Technology, 15, 4106-4112.
29. Zhang, M., Zhou, H., Liu, L., Feng, L., Yang, J., Wang, G., & Zhong, N. (2018). Randomized EEG functional brain networks in major depressive disorders with greater resilience and lower rich-club coefficient. Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology, 129 (4), 743-758.
30. Shim, M., Im, C. H., Kim, Y. W., & Lee, S. H. (2018). Altered cortical functional network in major depressive disorder: A resting-state electroencephalogram study. NeuroImage. Clinical, 19, 1000-1007.
31. Iandolo, R., Semprini, M., Buccelli, S., Barban, F., Zhao, M., Samogin, J., Bonassi, G., Avanzino, L., Mantini, D., & Chiappalone, M. (2020). Small-world propensity reveals the frequency specificity of resting state networks. IEEE Open Journal of Engineering in Medicine and Biology, 1, 57-64.
32. Luo, Q., Deng, Z., Qin, J., Wei, D., Cun, L., Qiu, J., Hitchman, G., & Xie, P. (2015). Frequency dependant topological alterations of intrinsic functional connectome in major depressive disorder. Scientific Reports, 5, 9710.
33. Singh, M. K., Kesler, S. R., Hosseini, S. H., Kelley, R. G., Amatya, D., Hamilton, J. P., Chen, M. C., & Gotlib, I. H. (2013). Anomalous gray matter structural networks in major depressive disorder. Biological Psychiatry, 74(10), 777-785.
34. Ajilore, O., Lamar, M., & Kumar, A. (2014). Association of brain network efficiency with aging, depression, and cognition. The American Journal of Geriatric Psychiatry, 22(2), 102-110.
35. Guo, H., Cheng, C., Cao, X., Xiang, J., Chen, J., & Zhang, K. (2014). Resting-state functional connectivity abnormalities in first-onset unmedicated depression. Neural Regeneration Research, 9(2), 153-163.
36. Li, H., Zhou, H., Yang, Y., Wang, H., & Zhong, N. (2017). More randomized and resilient in the topological properties of functional brain networks in patients with major depressive disorder. Journal of Clinical Neuroscience, 44, 274-278.
37. Wang, T., Wang, K., Qu, H., Zhou, J., Li, Q., Deng, Z., Du, X., Lv, F., Ren, G., Guo, J., Qiu, J., & Xie, P. (2016). Disorganized cortical thickness covariance network in major depressive disorder implicated by aberrant hubs in large-scale networks. Scientific Reports, 6(1), 1-12.
38. Borchardt, V., Lord, A. R., Li, M., van der Meer, J., Heinze, H. J., Bogerts, B., Breakspear, M., & Walter, M. (2016). Preprocessing strategy influences graph‐based exploration of altered functional networks in major depression. Human Brain Mapping, 37(4), 1422-1442.
39. Hosseinifard, B., Moradi, M. H., & Rostami, R. (2013). Classifying depression patients and normal subjects using machine learning techniques and nonlinear features from EEG signal. Computer methods and programs in biomedicine, 109, 339-345.
40. Mumtaz, W., Xia, L., Ali, S. S. A., Yasin, M. A. M., Hussain, M., & Malik, A. S. (2017). Electroencephalogram (EEG)-based computer-aided technique to diagnose major depressive disorder (MDD). Biomed Signal Process Control, 31, 108-115.
41. Li, X. W., Zhang, X., Zhu, J., Mao, W. D., Sun, S. T., Wang, Z. H., Xia, C., & Hu, B. (2019). Depression Recognition Using Machine Learning Methods with Different Feature Generation Strategies. Artificial Intelligence in Medicine, 99, 13-24.
42. Ding, X., Yue, X., Zheng, R., Bi, C., Li, D., & Yao, G. (2019). Classifying major depression patients and healthy controls using EEG, eye tracking and galvanic skin response data. Journal of affective disorders, 251, 156-161.
43. Acharya, U. R., Oh, S. L., Hagiwara, Y., Tan, J. H., Adeli, H., & Subha, D. P. (2018). Automated EEG-based screening of depression using deep convolutional neural network. Computer methods and programs in biomedicine, 161, 103-113.
44. Mumtaz, W., & Qayyum, A. (2019). A deep learning framework for automatic diagnosis of unipolar depression. International journal of medical informatics, 132, 103983.
45. Aditi, S., Pradeep, T., Harshit, B., Divya, A., & Arpit, B. (2021). A LSTM based deep learning network for recognizing emotions using wireless brainwave driven system. Expert Systems with Applications, 173, 114516.
46. Bailey, N. W., Hoy, K. E., Rogasch, N. C., Thomson, R. H, McQueen, S., & Elliot, D. (2019). Differentiating responders and non-responders to rTMS treatment for depression after one week using resting EEG connectivity measures. Journal of affective disorders, 242, 68-79.
47. Hasanzadeh, F., Mohebbi, M., & Rostami, R. (2019). Prediction of rTMS treatment response in major depressive disorder using machine learning techniques and nonlinear features of EEG signal. Journal of affective disorders, 256, 132-42.
48. Al-Kaysi, A. M., Al-Ani, A., Loo, C. K., Powell, T. Y., Martin, D. M., Breakspear, M., & Boonstra, T. W. (2017). Predicting tDCS treatment outcomes of patients with major depressive disorder using automated EEG classification. Journal of affective disorders, 208, 597-603.
49. Khodayari-Rostamabad, A., Reilly, J. P., Hasey, G. M., de Bruin, H., & Maccrimmon, D. J. (2013). A machine learning approach using EEG data to predict response to SSRI treatment for major depressive disorder. Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology, 124(10), 1975-1985.
50. Ghahramani, Z., & Hinton, G. E. (1997). The EM Algorithm for Mixtures of Factor Analyzers.
51. Shim, M., Jin, M. J., Im, C. H., & Lee, S. H. (2019). Machine-learning-based classification between post-traumatic stress disorder and major depressive disorder using P300 features. NeuroImage. Clinical, 24, 102001.
52. Zandvakili, A., Philip, N. S., Jones, S. R., Tyrka, A. R., Greenberg, B. D., & Carpenter, L. L. (2019). Use of machine learning in predicting clinical response to transcranial magnetic stimulation in comorbid posttraumatic stress disorder and major depression: A resting state electroencephalography study. Journal of affective disorders, 252, 47-54.
53. Sung, J. L., Guo, L. Y., Liu, C. H., Lee, P., Yen, C. W., & Liaw, L. J. (2021). Characterizing human postural stability by using features extracted from the correlation structure of ground reaction force components. IEEE Access, 9, 77398-77406.
54. Smarr, K. L., & Keefer, A. L. (2011). Measures of depression and depressive symptoms: Beck Depression Inventory‐II (BDI‐II), Center for Epidemiologic Studies Depression Scale (CES‐D), Geriatric Depression Scale (GDS), Hospital Anxiety and Depression Scale (HADS), and Patient Health Questionnaire‐9 (PHQ‐9). Arthritis Care & Research, 63(11), 454-466.
55. Zhang, M., Kim, J. C., Li, Y., Shapiro, B. B., Porszasz, J., Bross, R., Feroze, U., Upreti, R., Martin, D., Kalantar-Zadeh, K., & Kopple, J. D. (2014). Relation between anxiety, depression, and physical activity and performance in maintenance hemodialysis patients. Journal of Renal Nutrition, 24(4), 252-260.
56. Debener, S., Beauducel, A., Nessler, D., Brocke, B., Heilemann, H., & kayser, J. (2000). Is resting anterior EEG alpha asymmetry a trait marker for depression? Neuropsychobiology, 41(1), 31-37.
57. 林宜美（2018）。合併/未合併焦慮症狀的憂鬱症之量化腦波暨神經回饋療效研究。105年科技部研究計劃結案報告。
58. 林宜美（2019）。合併/未合併焦慮症狀的憂鬱症之量化腦波暨神經回饋療效研究-II。106年科技部研究計劃結案報告。
59. 陳心怡（2000）。貝克憂鬱量表第二版(中文版)指導手冊。台北市：中國行為科學社。
60. 林一真（2000）。貝克焦慮量表指導手冊。台北市：中國行為科學社。
61. Buysse, D. J., Reynolds III, C. F., Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193-213.
62. Nicassio, P. M., Mendlowitz, D. R., Fussell, J. J., & Petras, L. (1985). The phenomenology of the pre-sleep state: The development of the pre-sleep arousal scale. Behaviour Research and Therapy, 23(3), 263-271.
63. 林宜美（2013）。探討憂鬱症患者之心理病理機制：自主神經失調與心臟迷走神經調控模式。101年國科會計畫研究計劃結案報告。
64. BrainMaster Technologies. Devices/Hardware & Software. Retrieved from https://brainmaster.com/
65. 白映俞(2019)。大腦解剖秘密。取自照護線上Care online網址https://www.careonline.com.tw/2019/01/brain.html
66. 胡欣恬（2018）。認識大腦的CEO（大腦皮質篇）。取自財團法人光寶文教基金會網址http://www.liteoncf.org.tw/dm/139/139.html
67. Misulis, K. E. (2013). Atlas of EEG, Seizure Semiology, and Management. New York, NY: Oxford University Press.
68. Collura, T. F. (2014). Technical foundations of neurofeedback. New York, NY: Routledge.
69. Benesty, P. D. J., Chen, J., Huang, Y., & Cohen, P. I. (2009, March 5). Pearson correlation coefficient. in Noise Reduction in Speech Processing, Berlin, Germany: Springer, 1-4.
70. Bartlett, M. S. (1950). Periodogram analysis and continuous spectra. Biometrika, 37, 1-16.
71. Welch, P. D. (1967). The use of Fast Fourier Transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Transactions on Audio and Electroacoustics, 15 (2): 70-73.
72. Heideman, M. T., Johnson, D. H., & Burrus, C. S. (1984). Gauss and the history of the fast Fourier transform. IEEE ASSP Magazine, 1(4), 14-21.
73. Abdi, H., & Williams, L. J. (2010). Principal component analysis. Wiley Interdisciplinary Reviews: Computational Statistics, 2, 433-459.
74. Mehta, M. L., & Dyson, F. J. (1962). Statistical theory of the energy levels of complex systems. I, II, III. Journal of Mathematical Physics, 3, 140-175.
75. Mehta, M. L., & Dyson, F. J. (1963). Statistical theory of the energy levels of complex systems. IV. Journal of Mathematical Physics, 4, 701-712.
76. Mehta, M. L., & Dyson, F. J. (1963). Statistical theory of the energy levels of complex systems. V. Journal of Mathematical Physics, 4, 713-719.
77. Wigner, E. P. (1951). On a Class of Analytic Functions from the Quantum Theory of Collisions. Ann. Math, 36-67.
78. Wigner, E. P. (1951). On the statistical distribution of the widths and spacings of nuclear resonance levels. Mathematical Proceedings of the Cambridge Philosophical Society, 790-798.
79. Shannon, C. E. (1948). A Mathematical Theory of Communication. Bell System Technical Journal, 27(4), 623-656.
80. Rubinov, M., & Sporns, O. (2010). Complex network measures of brain connectivity: uses and interpretations. Neuroimage, 52, 1059-1069.
81. Wikipedia, the free encyclopedia. Simplicial complex. Retrieved from https://en.wikipedia.org/wiki/Simplicial_complex#/media/File:Simplicial_complex_example.svg/
82. Ioannides, A. A. (2007). Dynamic functional connectivity. Current Opinion in Neurobiology, 17, 161-170.
83. Wang, J., Wang, L., Zang, Y., Yang, H., Tang, H., Gong, Q., Chen, Z., Zhu, C., & He, Y. (2009). Parcellation-dependent small-world brain functional networks: a resting-state fMRI study. Human brain mapping, 30(5), 1511-1523.
84. Honey, C. J., Sporns, O., Cammoun, L., Gigandet, X., Thiran, J. P., Meuli, R., & Hagmann, P. (2009). Predicting human resting-state functional connectivity from structural connectivity. Proceedings of the National Academy of Sciences of the United States of America, 106(6), 2035-2040.
85. Saramäki, J., Kivelä, M., Onnela, J. P., Kaski, K., & Kertész, J. (2007). Generalizations of the clustering coefficient to weighted complex networks. Physical review. E, Statistical, nonlinear, and soft matter physics, 75, 027105.
86. Honey, C. J., Kötter, R., Breakspear, M., & Sporns, O. (2007). Network structure of cerebral cortex shapes functional connectivity on multiple time scales. Proceedings of the National Academy of Sciences of the United States of America, 104(24), 10240-10245.
87. Dijkstra, E. W. (1959). A note on two problems in connexion with graphs. Numerische Mathematik, 1, 269-271.
88. IT邦幫忙網路作者ramonliao（2019）。最短路徑 (Dijkstra 演算法)。取自IT邦幫忙網址https://ithelp.ithome.com.tw/articles/10209593
89. Sporns O. (2013). Structure and function of complex brain networks. Dialogues in clinical neuroscience, 15(3), 247-262.
90. Onnela, J. P., Saramäki, J., Kertész, J., & Kaski, K. (2005). Intensity and coherence of motifs in weighted complex networks. Physical review. E, Statistical, nonlinear, and soft matter physics, 71, 065103.
91. Fagiolo G. (2007). Clustering in complex directed networks. Physical review. E, Statistical, nonlinear, and soft matter physics, 76, 026107.
92. Latora, V., & Marchiori, M. (2001). Efficient behavior of small-world networks. Physical review letters, 87(19), 198701.
93. Achard, S., & Bullmore, E. (2007). Efficiency and cost of economical brain functional networks. PLoS computational biology, 3(2), e17.
94. OOSGA團隊（2021）。機器學習－定義、演算法、以及商業應用。取自OOSGA官方網址https://oosga.com/pillars/machine-learning/
95. Solomatine, D. P., & Shrestha, D. L. (2004). AdaBoost. RT: a boosting algorithm for regression problems. IEEE International Joint Conference, 2, 1163-1168.
96. Ho, T. K. (1998). The Random Subspace Method for Constructing Decision Forests. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(8), 832-844.
97. Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794.
98. Kursa, M. B., & Rudnicki, W. R. (2010). Feature Selection with the Boruta Package. Journal of Statistical Software, 36, 11.
99. Zabihi, M., Kiranyaz, S., & Gabbouj, M. (2019, September). Sepsis prediction in intensive care unit using ensemble of XGboost models. 2019 Computing in Cardiology Conference (CinC) at the meeting of Singapore, Singapore.
100. Kohavi, R. (1995). A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, 2 (12), 1137-1143.
101. Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 37-46.