注意力不足過動症是一種常好發於孩童及青少年的精神疾病，主要症狀包含容易分心及無法專注在特定的事物上，同時也對患者的日常生活造成極大的影響，然而這些源自於異常的腦部活動所造成的症狀，其實也是近年來研究的主要方向之一，在功能性磁振造影的技術日益成熟下，我們能夠透過更完善的腦部神經成像來辨識及分析過動症的腦部訊號，同時透過深度學習的方式來加強辨識的結果，在這項基礎上，我們從兩個不同的角度提出一個可以改善過動症在功能性磁振造影中的辨識結果，其一，是整合每一筆資料的個人特質(例如:年齡、性別)，來給予我們的模型額外的資訊進行辨識，其二，是增加訓練的資料，對我們的資料集進行資料增強的方法，將額外的資料集進行轉化合併後，進而加強我們的模型表現，在實驗一中我們提出了以表型特徵計算為主的一種注意力機制，並結合在腦部嵌入式特徵模型中做辨識，而實驗二則是透過訓練一個轉化自閉症資料集的模型，並將轉化後的假資料合併至原本的訓練集，兩項實驗在ADHD-200這份資料集中都得到很顯著的改善，並結合這兩項實驗，可以在改善過動症的預測方面有很明顯的進步。

Attention Deficit/Hyperactivity Disorder is one of the common disease prevalent in adolescents and child. Main symptoms include impulsiveness, distractibility, and deficient concentration which causes delayed social relationship. However, these symptoms caused by abnormal brain activity are actually one of the main research directions in recent years. Within the technology of functional magnetic resonance imaging, we are able to identify and analyze brain signal in ADHD. At the same time, deep learning algorithm also enhance the classification performance. On this basis, we propose two algorithm strategy from different kinds of aspect to improve ADHD classification in fMRI. First, integrating personal attribute to our modality in order to obtain extra reliable information to assist the classification. Second, we implement a data augmentation to increase the amount of training data also enhance our model’s performance. Overall, we propose a CVAE-based brain embedding network with an attention mechanism derived from phenotypic attributes and data augmentation strategy.
Summarizing all the strategy we propose demonstrated a significant improvement in ADHD classification.

誌謝 I
摘要 II
ABSTRACT III
CONTENTS IV
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 DATABASE AND FEATURES 6
2.1 DATABASE 6
2.1.1 ADHD-200 6
2.1.2 ABIDE-II 7
2.2 DATA PREPROCESSING 8
2.2 FEATURES 10
CHAPTER 3 ATTENTION ATTRIBUTE-ENHANCED NETWORK 11
3.1 RESEARCH M ETHODOLOGY 12
3.1.1 Cvae Based Brain Embedding Network 12
3.1.2 Phenotypic Attribute Attention Mechanism 13
3.1.3 Support Vector Machine 14
3.2 EXPERIMENT & RESULTS 14
3.2.1 Modality Comparison 17
3.2.2 Experiment Result 11
3.3 ATTENTION ANALYSIS & VISUALIZATION 20
3.3.1 Primary Group Analysis 21
3.3.2 Group Analysis with Gender 22
3.3.3 Group Analysis with Age 24
3.3.4 ADHD Subtype Analysis 25
CHAPTER 4 EXPERIMENT OF DATA AUGMENTATION. 27
4.1 RESEARCH METHODOLOGY 28
4.1.1 Cycle Generative Adversarial Network 28
4.1.2 Phenotypic Attributes Assignment 29
4.2 EXPERIMENT & RESULTS 30
4.3 RESULT ANALYSIS & VISUALIZATION 33
4.3.1 Augmentation Visualization 33
4.3.2 Connectivity Variation 35
4.3.3 Attention Analysis & Visualization 36

CHAPTER 5 CONCLUSION 39
REFERENCE 41

• [1] National Collaborating Centre for Mental Health (UK. Attention deficit hyperactivity disorder: diagnosis and management of ADHD in children, young people and adults. British Psychological Society (UK), 2009.
• [2] Cuffe, Steven P., Charity G. Moore, and Robert E. McKeown. "Prevalence and correlates of ADHD symptoms in the national health interview survey." Journal of attention disorders 9.2 (2005): 392-401.
• [3] Lifford, Kate J., Gordon T. Harold, and Anita Thapar. "Parent–child relationships and ADHD symptoms: a longitudinal analysis." Journal of abnormal child psychology 36.2 (2008): 285-296.
• [4] Castellanos, F. Xavier, et al. "Characterizing cognition in ADHD: beyond executive dysfunction." Trends in cognitive sciences 10.3 (2006): 117-123.
• [5] Liotti, Mario, et al. "Abnormal brain activity related to performance monitoring and error detection in children with ADHD." Cortex 41.3 (2005): 377-388.
• [6] S Konrad, Kerstin, et al. "Dysfunctional attentional networks in children with attention deficit/hyperactivity disorder: evidence from an event-related functional magnetic resonance imaging study." Biological psychiatry 59.7 (2006): 643-651.
• [7] J Rubia, Katya, et al. "Impulsiveness as a timing disturbance: neurocognitive abnormalities in attention-deficit hyperactivity disorder during temporal processes and normalization with methylphenidate." Philosophical Transactions of the Royal Society B: Biological Sciences 364.1525 (2009): 1919-1931.
• [8] Smith, Anna B., et al. "Reduced activation in right lateral prefrontal cortex and anterior cingulate gyrus in medication‐naïve adolescents with attention deficit hyperactivity disorder during time discrimination." Journal of Child Psychology and Psychiatry 49.9 (2008): 977-985.
• [9] Zou, Liang, et al. "3D CNN based automatic diagnosis of attention deficit hyperactivity disorder using functional and structural MRI." IEEE Access 5 (2017): 23626-23636.
• [10] Itani, Sarah, Fabian Lecron, and Philippe Fortemps. "A multi-level classification framework for multi-site medical data: Application to the ADHD-200 collection." Expert Systems with Applications 91 (2018): 36-45.
• [11] Giedd, Jay N., et al. "Brain development during childhood and adolescence: a longitudinal MRI study." Nature neuroscience 2.10 (1999): 861-863..
• [12] Thompson, Ross A., and Charles A. Nelson. "Developmental science and the media: Early brain development." American Psychologist 56.1 (2001): 5.
• [13] Haier, Richard J., et al. "The neuroanatomy of general intelligence: sex matters." NeuroImage 25.1 (2005): 320-327.
• [14] Krain, Amy L., and F. Xavier Castellanos. "Brain development and ADHD." Clinical psychology review 26.4 (2006): 433-444.
• [15] Zhang, Chao, et al. "Sex and age effects of functional connectivity in early adulthood." Brain connectivity 6.9 (2016): 700-713.
• [16] Miao, B., et al. "Classification of ADHD individuals and neurotypicals using reliable RELIEF: A resting-state study." IEEE Access 7 (2019): 62163-62171.
• [17] Vaswani, Ashish, et al. "Attention is all you need." Advances in neural information processing systems. 2017.
• [18] Tay, Chiat-Pin, Sharmili Roy, and Kim-Hui Yap. "Aanet: Attribute attention network for person re-identifications." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019.
• [19] Johnson, Mark H., et al. "Annual Research Review: Infant development, autism, and ADHD–early pathways to emerging disorders." Journal of Child Psychology and Psychiatry 56.3 (2015): 228-247.
• [20] Mayes, Susan Dickerson, et al. "Autism and ADHD: Overlapping and discriminating symptoms." Research in Autism Spectrum Disorders 6.1 (2012): 277-285.
•
• [21] Chantiluke, Kaylita, et al. "Disorder-specific functional abnormalities during temporal discounting in youth with Attention Deficit Hyperactivity Disorder (ADHD), Autism and comorbid ADHD and Autism." Psychiatry Research: Neuroimaging 223.2 (2014): 113-120.
• [22] Bellec, Pierre, et al. "The neuro bureau ADHD-200 preprocessed repository." Neuroimage 144 (2017): 275-286.
• [23] Craddock, Cameron, et al. "The neuro bureau preprocessing initiative: open sharing of preprocessed neuroimaging data and derivatives." Neuroinformatics 4 (2013).
• [24] Penny, William D., et al., eds. Statistical parametric mapping: the analysis of functional brain images. Elsevier, 2011.
• [25] Tzourio-Mazoyer, Nathalie, et al. "Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain." Neuroimage 15.1 (2002): 273-289.
• [26] Kingma, Diederik P., and Max Welling. "Auto-encoding variational bayes." arXiv preprint arXiv:1312.6114 (2013).
• [27] Choi, Hongyoon. "Functional connectivity patterns of autism spectrum disorder identified by deep feature learning." arXiv preprint arXiv:1707.07932 (2017).
• [28] Bar-Hillel, Aharon, Hertz, Tomer, Shental, Noam, and Weinshall, Daphna. Learning a mahalanobis metric from equivalence constraints. JMLR, 2005.
• [29] Lin, Yi-Lin, and Gang Wei. "Speech emotion recognition based on HMM and SVM." 2005 international conference on machine learning and cybernetics. Vol. 8. IEEE, 2005.
• [30] Riaz, Atif, et al. "Fusion of fMRI and non-imaging data for ADHD classification." Computerized Medical Imaging and Graphics 65 (2018): 115-128.
• [31] Itani, Sarah, Fabian Lecron, and Philippe Fortemps. "A multi-level classification framework for multi-site medical data: Application to the ADHD-200 collection." Expert Systems with Applications 91 (2018): 36-45.
• [32] van der Stoep, Nathan, Albert Postma, and Tanja CW Nijboer. "Multisensory perception and the coding of space." (2017).
• [33] Mechelli, Andrea, et al. "Differential effects of word length and visual contrast in the fusiform and lingual gyri during." Proceedings of the Royal Society of London. Series B: Biological Sciences 267.1455 (2000): 1909-1913.
• [34] Cavanna, Andrea E., and Michael R. Trimble. "The precuneus: a review of its functional anatomy and behavioural correlates." Brain 129.3 (2006): 564-583.
• [35] Qiu, Ming-guo, et al. "Changes of brain structure and function in ADHD children." Brain topography 24.3-4 (2011): 243-252.
• [36] Uddin, Lucina Q., et al. "Network homogeneity reveals decreased integrity of default-mode network in ADHD." Journal of neuroscience methods 169.1 (2008): 249-254.
• [37] Chiang, Ching-Tai, et al. "Increased Temporal Lobe Beta Activity in Boys With Attention-Deficit Hyperactivity Disorder by LORETA Analysis." Frontiers in Behavioral Neuroscience 14 (2020): 85.
• [38] Park, Bo-yong, and Hyunjin Park. "Connectivity differences between adult male and female patients with attention deficit hyperactivity disorder according to resting-state functional MRI." Neural regeneration research 11.1 (2016): 119.
• [39] McLeod, Kevin R., et al. "Functional connectivity of neural motor networks is disrupted in children with developmental coordination disorder and attention-deficit/hyperactivity disorder." NeuroImage: Clinical 4 (2014): 566-575.
• [40] Schapiro, Mark B., et al. "BOLD-fMRI signal increases with age in selected brain regions in children." Neuroreport 15.17 (2004): 2575.

• [41] Solanto, Mary V., et al. "Social functioning in predominantly inattentive and combined subtypes of children with ADHD." Journal of attention disorders 13.1 (2009): 27-35.
• [42] Zhu, Jun-Yan, et al. “Unpaired image-to-image translation using cycle-consistent adversarial networks." Proceedings of the IEEE international conference on computer vision. 2017
• [43] Engin, Deniz, Anil Genç, and Hazim Kemal Ekenel. "Cycle-dehaze: Enhanced cyclegan for single image dehazing." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2018.
• [44] Larsen, Anders Boesen Lindbo, et al. "Autoencoding beyond pixels using a learned similarity metric." International conference on machine learning. PMLR, 2016.
• [45] Larsen, Anders Boesen Lindbo, et al. "Autoencoding beyond pixels using a learned similarity metric." International conference on machine learning. PMLR, 2016.
• [45] Mostofsky, Stewart H., et al. "Decreased connectivity and cerebellar activity in autism during motor task performance." Brain 132.9 (2009): 2413-2425.
• [46] Xu, Jinping, et al. "Specific functional connectivity patterns of middle temporal gyrus subregions in children and adults with autism spectrum disorder." Autism Research 13.3 (2020): 410-422.
• [47] Jou, Roger J., et al. "Enlarged right superior temporal gyrus in children and adolescents with autism." Brain research 1360 (2010): 205-212.
• [48] Sato, João Ricardo, et al. "Measuring network's entropy in ADHD: A new approach to investigate neuropsychiatric disorders." Neuroimage 77 (2013): 44-51.
• [49] Kipf, Thomas N., and Max Welling. "Semi-supervised classification with graph convolutional networks." arXiv preprint arXiv:1609.02907 (2016).