憂鬱症目前為盛行率較高的精神疾病。臨床上，醫師會根據患者的病徵、病史及病患過去一周的個案自述來進行評估，若患者處於憂鬱狀態長達兩周以上，則能確診為憂鬱症。此外，醫師還會藉由不同的憂鬱症篩檢量表進行檢測，提高確診的可信度，並且在經過審慎評估後，依其病症的嚴重程度，採取不同的治療方式。但是目前對於憂鬱症的診斷標準較為主觀，缺乏科學方法進行客觀的檢驗，因此本研究嘗試運用深度學習技術建立一套診斷系統，希望能透過受試者的功能性核磁共振造影(functional Magnetic Resonance Imaging, fMRI)配合交叉樣本熵演算法，先行判斷患者是否真的罹患憂鬱症，隨後再預估患者的病症嚴重程度。憂鬱症判別部分，除了卷積神經網路模型外，我們同時使用自動編碼器與支援向量機輔助，使得分類準確率達到80%以上；而病症嚴重程度評估方面，藉由分類模型與迴歸模型的合併使用，在測試集的均方根誤差為3.52，若再加以改善的話，效果會優於僅使用單一迴歸模型進行預測的結果。

Major depressive disorder is a mental disorder with a high prevalence. Clinically, the diagnosis of depression is based on the person's symptoms, biographical history, and reported experiences. Moreover, if a person is in a state of depression for more than two weeks, the diagnosis can be made. Besides, several rating scales would be used in mental status examination to evaluate the severity of depression more thoroughly. According to the severity of the disease, different types of treatment would be used. However, the current diagnostic criteria for depression are of concern because it relies on subjective interpretation and lack of scientific methods for confirming. Hence, we attempt to build a deep learning system that could classify the patients from normal and evaluate the severity of depression using resting-state functional magnetic resonance imaging (fMRI) with cross-sample entropy. For the classification of depression, we use autoencoder and support vector machines better to make the convolutional neural network model’s performance better, and the combining of classification model and a regression model was used for scale prediction. From the result, the accuracy of the depression classification was more than 80%, and the root means square error for the scale prediction during testing was 3.52. The performance would be better if the model in the classification stage of scale prediction was improved by comparing it to those of a scheme only with a regression stage.

摘要
Abstract
誌謝
目錄
圖目錄
表目錄
第一章-------------------1
第二章-------------------6
第三章-------------------10
第四章-------------------18
第五章-------------------26
第六章-------------------33
參考文獻-----------------34

[1] D. J. Kupfer, E. Frank, and M. L. Phillips, "Major depressive disorder: new clinical, neurobiological, and treatment perspectives," Focus, vol. 14, no. 2, pp. 266-276, 2016.
[2] W. H. Organization, "Depression," January 30 2020.
[3] NIMH, "Depression," May 2016.
[4] J. Battle, "Relationship between self-esteem and depression," Psychological Reports, vol. 42, no. 3, pp. 745-746, 1978.
[5] S. Cavanaugh, D. C. Clark, and R. D. Gibbons, "Diagnosing depression in the hospitalized medically ill," Psychosomatics, vol. 24, no. 9, pp. 809-815, 1983.
[6] N. Tsuno, A. Besset, and K. Ritchie, "Sleep and depression," The Journal of clinical psychiatry, 2005.
[7] S. Bachmann, "Epidemiology of suicide and the psychiatric perspective," International journal of environmental research and public health, vol. 15, no. 7, p. 1425, 2018.
[8] M. aan het Rot, S. J. Mathew, and D. S. Charney, "Neurobiological mechanisms in major depressive disorder," Cmaj, vol. 180, no. 3, pp. 305-313, 2009.
[9] L. L. Patton and M. Glick, The ADA practical guide to patients with medical conditions. Wiley Online Library, 2012.
[10] M. Hamilton and W. Guy, "Hamilton depression scale," Group, vol. 1, p. 4, 1976.
[11] A. T. Beck, R. A. Steer, and G. K. Brown, Beck depression inventory (BDI-II). Pearson, 1996.
[12] D. Watson, L. A. Clark, and A. Tellegen, "Development and validation of brief measures of positive and negative affect: the PANAS scales," Journal of personality and social psychology, vol. 54, no. 6, p. 1063, 1988.
[13] L. K. Sharp and M. S. Lipsky, "Screening for depression across the lifespan," Am Fam Physician, vol. 66, pp. 1001-1008, 2002.
[14] V. Patel, R. Araya, and P. Bolton, "Treating depression in the developing world," Tropical Medicine & International Health, vol. 9, no. 5, pp. 539-541, 2004.
[15] N. C. C. f. M. Health, "Depression: the treatment and management of depression in adults (updated edition)," 2010: British Psychological Society.
[16] A. C. Yang et al., "Complexity of spontaneous BOLD activity in default mode network is correlated with cognitive function in normal male elderly: a multiscale entropy analysis," Neurobiology of Aging, vol. 34, no. 2, pp. 428-438, 2013.
[17] M. O. Sokunbi, W. Fung, V. Sawlani, S. Choppin, D. E. Linden, and J. Thome, "Resting state fMRI entropy probes complexity of brain activity in adults with ADHD," Psychiatry Research: Neuroimaging, vol. 214, no. 3, pp. 341-348, 2013.
[18] M. O. Sokunbi et al., "Nonlinear complexity analysis of brain FMRI signals in schizophrenia," Plos one, vol. 9, no. 5, p. e95146, 2014.
[19] Y. Niu et al., "Dynamic Complexity of Spontaneous BOLD Activity in Alzheimer's Disease and Mild Cognitive Impairment Using Multiscale Entropy analysis," Frontiers in neuroscience, vol. 12, p. 677, 2018.
[20] P.-S. Ho et al., "Complexity analysis of resting state fMRI signals in depressive patients," in 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2017, pp. 3190-3193: IEEE.
[21] R. Bhaumik et al., "Multivariate pattern analysis strategies in detection of remitted major depressive disorder using resting state functional connectivity," NeuroImage: Clinical, vol. 16, pp. 390-398, 2017.
[22] J. R. Sato, J. Moll, S. Green, J. F. Deakin, C. E. Thomaz, and R. Zahn, "Machine learning algorithm accurately detects fMRI signature of vulnerability to major depression," Psychiatry Research: Neuroimaging, vol. 233, no. 2, pp. 289-291, 2015.
[23] C. Bürger et al., "Differential abnormal pattern of anterior cingulate gyrus activation in unipolar and bipolar depression: an fMRI and pattern classification approach," Neuropsychopharmacology, vol. 42, no. 7, pp. 1399-1408, 2017.
[24] K. Yoshida et al., "Prediction of clinical depression scores and detection of changes in whole-brain using resting-state functional MRI data with partial least squares regression," PloS one, vol. 12, no. 7, p. e0179638, 2017.
[25] A. Payan and G. Montana, "Predicting Alzheimer's disease: a neuroimaging study with 3D convolutional neural networks," arXiv preprint arXiv:1502.02506, 2015.
[26] L. Zou, J. Zheng, C. Miao, M. J. Mckeown, and Z. J. Wang, "3D CNN based automatic diagnosis of attention deficit hyperactivity disorder using functional and structural MRI," IEEE Access, vol. 5, pp. 23626-23636, 2017.
[27] M. Khosla, K. Jamison, A. Kuceyeski, and M. R. Sabuncu, "Ensemble learning with 3D convolutional neural networks for functional connectome-based prediction," NeuroImage, vol. 199, pp. 651-662, 2019.
[28] H. Li, T. D. Satterthwaite, and Y. Fan, "Brain age prediction based on resting-state functional connectivity patterns using convolutional neural networks," in 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 2018, pp. 101-104: IEEE.
[29] J. Ashburner et al., "SPM12 manual," p. 2464, 2014.
[30] X.-W. Song et al., "REST: a toolkit for resting-state functional magnetic resonance imaging data processing," PloS one, vol. 6, no. 9, p. e25031, 2011.
[31] R. Sladky, K. J. Friston, J. Tröstl, R. Cunnington, E. Moser, and C. J. N. Windischberger, "Slice-timing effects and their correction in functional MRI," NeuroImage, vol. 58, no. 2, pp. 588-594, 2011.
[32] R. Zafar, A. S. Malik, N. Kamel, and S. C. Dass, "Importance of realignment parameters in fMRI data analysis," in 2015 IEEE International Conference on Signal and Image Processing Applications (ICSIPA), 2015, pp. 546-550: IEEE.
[33] J. L. Lancaster et al., "Automated Talairach atlas labels for functional brain mapping," Human Brain Mapping, vol. 10, no. 3, pp. 120-131, 2000.
[34] D. L. Collins, C. J. Holmes, T. M. Peters, and A. C. J. H. b. m. Evans, "Automatic 3‐D model‐based neuroanatomical segmentation," Human Brain Mapping, vol. 3, no. 3, pp. 190-208, 1995.
[35] M. N. Hallquist, K. Hwang, and B. J. N. Luna, "The nuisance of nuisance regression: spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity," NeuroImage, vol. 82, pp. 208-225, 2013.
[36] E. T. Rolls, C.-C. Huang, C.-P. Lin, J. Feng, and M. Joliot, "Automated anatomical labelling atlas 3," NeuroImage, vol. 206, p. 116189, 2020.
[37] N. Tzourio-Mazoyer et al., "Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain," Neuroimage, vol. 15, no. 1, pp. 273-289, 2002.
[38] L.-S. Young, "Dimension, entropy and Lyapunov exponents," Ergodic theory and dynamical systems, vol. 2, no. 1, pp. 109-124, 1982.
[39] S. Pincus, "Approximate entropy (ApEn) as a complexity measure," Chaos: An Interdisciplinary Journal of Nonlinear Science, vol. 5, no. 1, pp. 110-117, 1995.
[40] J. S. Richman and J. R. Moorman, "Physiological time-series analysis using approximate entropy and sample entropy," American Journal of Physiology-Heart and Circulatory Physiology, vol. 278, no. 6, pp. H2039-H2049, 2000.
[41] T. Zhang, Z. Yang, and J. H. Coote, "Cross‐sample entropy statistic as a measure of complexity and regularity of renal sympathetic nerve activity in the rat," Experimental physiology, vol. 92, no. 4, pp. 659-669, 2007.
[42] 斎藤康毅, "Deep Learning：用Python進行深度學習的基礎理論實作," 2017. 歐萊禮
[43] Y. LeCun, Y. Bengio, and G. J. n. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436-444, 2015.
[44] A. Ng, "Sparse autoencoder," CS294A Lecture notes, vol. 72, no. 2011, pp. 1-19, 2011.
[45] C. Cortes and V. Vapnik, "Support-vector networks," Machine learning, vol. 20, no. 3, pp. 273-297, 1995.
[46] A. Patle and D. S. Chouhan, "SVM kernel functions for classification," in 2013 International Conference on Advances in Technology and Engineering (ICATE), 2013, pp. 1-9: IEEE.
[47] A. Gulli and S. Pal, Deep learning with Keras. Packt Publishing Ltd, 2017.
[48] M. D. J. a. e.-p. Zeiler, "ADADELTA: An Adaptive Learning Rate Method," p. arXiv:1212.5701Accessed on: December 01, 2012Available: https://ui.adsabs.harvard.edu/abs/2012arXiv1212.5701Z
[49] G.-Y. Chen et al., "Depression Scale Prediction with Cross-Sample Entropy and Deep Learning," in 2020 42nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2017, pp. 120-123: IEEE.
[50] N. U. Dosenbach, D. A. Fair, A. L. Cohen, B. L. Schlaggar, and S. E. Petersen, "A dual-networks architecture of top-down control," Trends in cognitive sciences, vol. 12, no. 3, pp. 99-105, 2008.