Publications-Theses
Article View/Open
Publication Export
-
題名 運用函數主成分分析於阿茲海默症之診斷
Application of functional principal component analysis to diagnosis of Alzheimer’s disease作者 李詠玄
Lee, Yong-Shiuan貢獻者 劉惠美
Liu, Hui-Mei
李詠玄
Lee, Yong-Shiuan關鍵詞 阿茲海默症
函數主成分分析
遞迴類神經網路
長短期記憶類神經網路
長期追蹤資料
Alzheimer’s disease
Functional principal component analysis
Recurrent neural networks
Long short-term memory networks
Longitudinal data日期 2022 上傳時間 1-Aug-2022 17:13:51 (UTC+8) 摘要 自二十世紀晚期對於探討阿茲海默症成因、病情發展與有效治療方式的研究大量增加。其中最重要的目標之一即為於早期診斷出阿茲海默症，也就是輕度認知障礙。診斷輕度認知障礙或阿茲海默症即是統計上的分類問題。通常阿茲海默症的相關研究資料皆為長期追蹤資料，由於資料收集的方式，使得資料多為稀疏性且不規則間隔的資料。再者，基於近年來醫學診斷工具，尤其是腦部顯影的技術進步與普及，阿茲海默症資料更常為具有高維度的資料。傳統上常用的統計分類方法對於此類資料型態有其侷限性。本研究首先將資料的變數視為只具有少數觀察值的函數，使用函數主成分分析工具來重建高維度、稀疏且不規則間隔的資料，使資料收集區間內的所有觀察時間點皆能有函數的估計值。接續再利用遞迴類神經網路中專門針對時間序列資料的長短期記憶類神經網路，來對研究對象做診斷的分類。本研究的實證結果指出在最佳情境下，此作法使用較多觀察值於訓練資料集，以及使用較多的輸入變數，能夠正確辨認出最多的早期輕度認知障礙者（十一個患者中正確辨認出五個）。顯示此法對於辨認早期的輕度認知障礙有較大的潛力。針對阿茲海默症此類醫學研究中常見的不平衡資料，未來可考慮加入重新採樣的方法或是成本考量的分類方法進一步發展優化本文所提出之程序。
Since the late 20th century, researches of Alzheimer’s disease intending to better understand the causes, the progression, and effective treatments of this disease have boosted. One of the most important purposes of these researches is to detect the disease at early stages, that is, the diagnosis of mild cognitive impairment. The diagnosis is certainly the classification problem in statistics. The research data of Alzheimer’s disease are usually longitudinal, which can be very sparse and irregularlyspaced as a result of data collection process. Additionally, the research data can also have high imensional features due to improvement in clinical neuroimaging techniques. Classical approaches for classification have limitations in using the sparse and irregular, highdimensional, longitudinal data. This study is the first to implement the tool of the functional principal component analysis to reconstruct the whole\nfunctions of all variables during the period, and then to apply the long shortterm memory networks, a recurrent neural network designed for time series data, for classification. The empirical results show that in the bestcase scenario this method identifies 5 out of 11 MCI cases in the testing dataset while the other methods only accurately predict 0 or 1 MCI case. The results suggest that this procedure has great potential for early detection of Alzheimer’s disease. The proposed method can further be developed for imbalanced data with resampling or costsensitive classification techniques.參考文獻 [1] M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat,\nG. Irving, M. Isard, et al. Tensorflow: A system for largescale\nmachine learning.\nIn 12th {USENIX} Symposium on Operating Systems Design and Implementation\n({OSDI} 16), pages 265–283, 2016.\n[2] A. Anoop, P. K. Singh, R. S. Jacob, and S. K. Maji. CSF biomarkers for Alzheimer’s\ndisease diagnosis. International journal of Alzheimer’s disease, 2010:Article ID\n606802, 12 pages, 2010.\n[3] A. Association. 2020 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia,\n16(3):391–460, 2020.\n[4] A. Association. 2021 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia,\n17(3):327–406, 2021.\n[5] S. Balakrishnan and D. Madigan. Decision trees for functional variables. In Sixth\nInternational Conference on Data Mining (ICDM’06), pages 798–802. IEEE, 2006.\n[6] E. Belli and S. Vantini. Measure inducing classification and regression trees for\nfunctional data. Statistical Analysis and Data Mining: The ASA Data Science Journal,\n2021.\n[7] Y. Bengio. Learning deep architectures for AI. Foundations and Trends in Signal\nProcessing, 2(1):1–127, 2009.\n[8] J. R. Berrendero, A. Justel, and M. Svarc. Principal components for multivariate\nfunctional data. Computational Statistics & Data Analysis, 55(9):2619–2634, 2011.\n[9] M. Bertoux, J. Lagarde, F. Corlier, L. Hamelin, J.F.\nMangin, O. Colliot, M. Chupin,\nM. N. Braskie, P. M. Thompson, M. Bottlaender, et al. Sulcal morphology in\nAlzheimer’s disease: An effective marker of diagnosis and cognition. Neurobiology\nof Aging, 84:41–49, 2019.\n[10] M. C. Biagioni and J. E. Galvin. Using biomarkers to improve detection of\nAlzheimer’s disease. Neurodegenerative Disease Management, 1(2):127–139,\n2011.\n[11] S. Borson, J. Scanlan, M. Brush, P. Vitaliano, and A. Dokmak. The MiniCog:\nA\ncognitive ‘vital signs’measure for dementia screening in multilingual\nelderly.\nInternational journal of geriatric psychiatry, 15(11):1021–1027, 2000.\n[12] S. Borson, J. M. Scanlan, P. Chen, and M. Ganguli. The MiniCog\nas a screen\nfor dementia: Validation in a populationbased\nsample. Journal of the American\nGeriatrics Society, 51(10):1451–1454, 2003.\n[13] L. Breiman. Bagging predictors. Machine Learning, 24(2):123–140, 1996.\n[14] L. Breiman. Random forests. Machine Learning, 45(1):5–32, 2001.\n[15] L. Breiman, J. H. Friedman, R. A. Olshen, and C. J. Stone. Classification and\nRegression Trees. Chapman & Hall/CRC, New York., 1984.\n[16] A. M. Brickman, J. J. Manly, L. S. Honig, D. Sanchez, D. ReyesDumeyer,\nR. A.\nLantigua, P. J. Lao, Y. Stern, J. P. Vonsattel, A. F. Teich, et al. Plasma ptau181,ptau217,\nand other bloodbased\nAlzheimer’s disease biomarkers in a multiethnic,\ncommunity study. Alzheimer’s & Dementia, 17(8):1353–1364, 2021.\n[17] R. S. Bucks, D. Ashworth, G. Wilcock, and K. Siegfried. Assessment of activities\nof daily living in dementia: Development of the bristol activities of daily living\nscale. Age and ageing, 25(2):113–120, 1996.\n[18] H. Buschke, G. Kuslansky, M. Katz, W. F. Stewart, M. J. Sliwinski, H. M. Eckholdt,\nand R. B. Lipton. Screening for dementia with the memory impairment screen.\nNeurology, 52(2):231–231, 1999.\n[19] B. D. Carpenter, C. Xiong, E. K. Porensky, M. M. Lee, P. J. Brown, M. Coats,\nD. Johnson, and J. C. Morris. Reaction to a dementia diagnosis in individuals with\nAlzheimer’s disease and mild cognitive impairment. Journal of the American Geriatrics\nSociety, 56(3):405–412, 2008.\n[20] L.H.\nChen and C.R.\nJiang. Multidimensional\nfunctional principal component\nanalysis. Statistics and Computing, 27(5):1181–1192, 2017.\n[21] W.C.\nCheng, L.H.\nChen, C.R.\nJiang, Y.M.\nDeng, D.W.\nWang, C.H.\nLin, R. Jou,\nJ.K.\nWang, and Y.L.\nWang. Sensible functional linear discriminant analysis effectively\ndiscriminates enhanced Raman spectra of Mycobacterium species. Analytical\nChemistry, 93(5):2785–2792, 2021. PMID: 33480698.\n[22] R. Chin, A. Ng, K. Narasimhalu, and N. Kandiah. Utility of the AD8 as a selfrating\ntool for cognitive impairment in an Asian population. American Journal of\nAlzheimer’s Disease & Other Dementias®, 28(3):284–288, 2013.\n[23] J.M.\nChiou, Y.T.\nChen, and Y.F.\nYang. Multivariate functional principal component\nanalysis: A normalization approach. Statistica Sinica, pages 1571–1596,\n2014.\n[24] K. Cho, B. Van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk,\nand Y. Bengio. Learning phrase representations using RNN encoderdecoder\nfor\nstatistical machine translation. In Proceedings of the 2014 Conference on Empirical\nMethods in Natural Language Processing (EMNLP), page 1724–1734. Association\nfor Computational Linguistics (ACL), Oct. 2014.\n[25] S. H. Cho, S. Woo, C. Kim, H. J. Kim, H. Jang, B. C. Kim, S. E. Kim, S. J. Kim, J. P.\nKim, Y. H. Jung, et al. Disease progression modelling from preclinical Alzheimer’\ns disease (AD) to AD dementia. Scientific reports, 11(1):1–10, 2021.\n[26] F. Chollet et al. Keras. urlhttps://github.com/fchollet/keras, 2015.\n[27] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio. Empirical evaluation of gated recurrent\nneural networks on sequence modeling. arXiv preprint arXiv:1412.3555,\n2014.\n[28] M. Conceição, A. KroneMartins,\nand A. da Silva. FPCA emulation of cosmological\nsimulations. In 2021 IEEE 17th International Conference on eScience\n(eScience), pages 225–226. IEEE, 2021.\n[29] C. Cortes and V. Vapnik. Support vector machine. Machine Learning, 20(3):273–\n297, 1995.\n[30] R. Cui, M. Liu, A. D. N. Initiative, et al. RNNbased\nlongitudinal analysis for\ndiagnosis of Alzheimer’s disease. Computerized Medical Imaging and Graphics,\n73:1–10, 2019.\n[31] J. M. Cuttler, E. Abdellah, Y. Goldberg, S. AlShamaa,\nS. P. Symons, S. E. Black,\nand M. Freedman. Low doses of ionizing radiation as a treatment for Alzheimer’\ns disease: A pilot study. Journal of Alzheimer’s Disease, 80(3):1119–1128, 2021.\n[32] A. Delaigle and P. Hall. Achieving near perfect classification for functional\ndata. Journal of the Royal Statistical Society: Series B (Statistical Methodology),\n74(2):267–286, 2012.\n[33] A. Delaigle and P. Hall. Classification using censored functional data. Journal of\nthe American Statistical Association, 108(504):1269–1283, 2013.\n[34] A. Delaigle, P. Hall, and N. Bathia. Componentwise classification and clustering\nof functional data. Biometrika, 99(2):299–313, 2012.\n[35] L. Deng and D. Yu. Deep learning: Methods and applications. Foundations and\nTrends in Signal Processing, 7(3–4):197–387, 2014.\n[36] B. Dunn, P. Stein, and P. Cavazzoni. Approval of Aducanumab for Alzheimer\ndisease—The FDA’s perspective. JAMA Internal Medicine, 181(10):1276–1278,\n2021.\n[37] S. ElSappagh,\nT. Abuhmed, S. R. Islam, and K. S. Kwak. Multimodal multitask\ndeep learning model for Alzheimer’s disease progression detection based on time\nseries data. Neurocomputing, 412:197–215, 2020.\n[38] A. Ezzati, M. J. Katz, A. R. Zammit, M. L. Lipton, M. E. Zimmerman, M. J. Sliwinski,\nand R. B. Lipton. Differential association of left and right hippocampal\nvolumes with verbal episodic and spatial memory in older adults. Neuropsychologia,\n93:380–385, 2016.\n[39] J. Fan and I. Gijbels. Local Polynomial Modelling and Its Applications. Chapman\n& Hall/CRC, London, 1996.\n[40] C. Feng, A. Elazab, P. Yang, T. Wang, F. Zhou, H. Hu, X. Xiao, and B. Lei. Deep\nlearning framework forAlzheimer’s disease diagnosis via 3DCNN\nand FSBiLSTM.\nIEEE Access, 7:63605–63618, 2019.\n[41] A. Field, J. Miles, and Z. Field. Discovering statistics using R. Sage Publications,\n2012.\n[42] M. F. Folstein, S. E. Folstein, and P. R. McHugh. “Minimental\nstate”: A practical\nmethod for grading the cognitive state of patients for the clinician. Journal of\npsychiatric research, 12(3):189–198, 1975.\n[43] P. Forouzannezhad, A. Abbaspour, C. Fang, M. Cabrerizo, D. Loewenstein,\nR. Duara, and M. Adjouadi. A survey on applications and analysis methods of\nfunctional magnetic resonance imaging for Alzheimer’s disease. Journal of neuroscience\nmethods, 317:121–140, 2019.\n[44] S. Förster, B. H. Yousefi, H.J.\nWester, E. Klupp, A. Rominger, H. Förstl, A. Kurz,\nT. Grimmer, and A. Drzezga. Quantitative longitudinal interrelationships between\nbrain metabolism and amyloid deposition during a 2year\nfollowup\nin patients with\nearly Alzheimer’s disease. European journal of nuclear medicine and molecular\nimaging, 39(12):1927–1936, 2012.\n[45] J. H. Friedman. Regularized discriminant analysis. Journal of the American Statistical\nAssociation, 84(405):165–175, 1989.\n[46] A. Gajardo, C. Carroll, Y. Chen, X. Dai, J. Fan, P. Z. Hadjipantelis, K. Han, H. Ji,\nH.G.\nMüller, and J.L.\nWang. fdapace: Functional Data Analysis and Empirical\nDynamics, 2021. R package version 0.5.7.\n[47] T. P. Garcia and K. Marder. Statistical approaches to longitudinal data analysis in\nneurodegenerative diseases: Huntington’s disease as a model. Current Neurology\nand Neuroscience Reports, 17(2):1–9, 2017.\n[48] S. Gauthier, P. RosaNeto,\nJ. A. Morais, C. Webster, et al. World Alzheimer report\n2021 Journey\nthrough the diagnosis of dementia. https://www.alzint.org/\nresource/world-alzheimer-report-2021/. Accessed: 20210928.\n[49] I. Gélinas, L. Gauthier, M. McIntyre, and S. Gauthier. Development of a functional\nmeasure for persons with Alzheimer’s disease: the disability assessment\nfor dementia. American Journal of Occupational Therapy, 53(5):471–481, 1999.\n[50] M. M. Ghazi, M. Nielsen, A. Pai, M. J. Cardoso, M. Modat, S. Ourselin,\nL. Sørensen, A. D. N. Initiative, et al. Training recurrent neural networks robust\nto incomplete data: Application to Alzheimer’s disease progression modeling.\nMedical Image Analysis, 53:39–46, 2019.\n[51] Y. Gupta, R. K. Lama, G.R.\nKwon, M. W. Weiner, P. Aisen, M. Weiner, R. Petersen,\nC. R. Jack Jr, W. Jagust, J. Q. Trojanowki, et al. Prediction and classification\nof Alzheimer’s disease based on combined features from apolipoproteinE\ngenotype,\ncerebrospinal fluid, MR, and FDGPET\nimaging biomarkers. Frontiers in\nComputational Neuroscience, 13:72, 2019.\n[52] Y. Gupta, K. H. Lee, K. Y. Choi, J. J. Lee, B. C. Kim, G. R. Kwon, N. R. C. for\nDementia, and A. D. N. Initiative. Early diagnosis of Alzheimer’s disease using\ncombined features from voxelbased\nmorphometry and cortical, subcortical, and\nhippocampus regions of MRI T1 brain images. PLoS One, 14(10):e0222446, 2019.\n[53] C. Happ and S. Greven. Multivariate functional principal component analysis for\ndata observed on different (dimensional) domains. Journal of the American Statistical\nAssociation, 113(522):649–659, 2018.\n[54] C. HappKurz.\nObjectoriented\nsoftware for functional data. Journal of Statistical\nSoftware, 93(5):1–38, 2020.\n[55] C. HappKurz.\nMFPCA: Multivariate Functional Principal Component Analysis\nfor Data Observed on Different Dimensional Domains, 2021. R package version\n1.39.\n[56] J. A. Hardy and G. A. Higgins. Alzheimer’s disease: The amyloid cascade hypothesis.\nScience, 256(5054):184–186, 1992.\n[57] K. Hasenstab, A. Scheffler, D. Telesca, C. A. Sugar, S. Jeste, C. DiStefano, and\nD. Şentürk. A multidimensional\nfunctional principal components analysis of EEG\ndata. Biometrics, 73(3):999–1009, 2017.\n[58] T. Hastie. [Flexible Parsimonious Smoothing and Additive Modeling]: Discussion.\nTechnometrics, 31(1):23–29, 1989.\n[59] T. Hastie, A. Buja, and R. Tibshirani. Penalized discriminant analysis. The Annals\nof Statistics, 23(1):73–102, 1995.\n[60] T. Hastie, R. Tibshirani, and A. Buja. Flexible discriminant analysis by optimal\nscoring. Journal of the American Statistical Association, 89(428):1255–1270,\n1994.\n[61] S. Hochreiter and J. Schmidhuber. Long shortterm\nmemory. Neural Computation,\n9(8):1735–1780, 1997.\n[62] H. Hodkinson. Evaluation of a mental test score for assessment of mental impairment\nin the elderly. Age and ageing, 1(4):233–238, 1972.\n[63] W. Huang, Y. Zhou, L. Tu, Z. Ba, J. Huang, N. Huang, and Y. Luo. TDP43:\nFrom\nAlzheimer’s disease to limbicpredominant\nagerelated\nTDP43\nencephalopathy.\nFrontiers in Molecular Neuroscience, 13:26, 2020.\n[64] S. Iddi, D. Li, P. S. Aisen, M. S. Rafii, W. K. Thompson, and M. C. Donohue.\nPredicting the course of Alzheimer’s progression. Brain Informatics, 6(1):1–18,\n2019.\n[65] S. Ioffe and C. Szegedy. Batch normalization: Accelerating deep network training\nby reducing internal covariate shift. In International Conference on Machine\nLearning, pages 448–456. PMLR, 2015.\n[66] Z. Ismail, L. AgüeraOrtiz,\nH. Brodaty, A. Cieslak, J. Cummings, C. E. Fischer,\nS. Gauthier, Y. E. Geda, N. Herrmann, J. Kanji, et al. The Mild Behavioral\nImpairment Checklist (MBIC):\nA rating scale for neuropsychiatric symptoms in\npredementia\npopulations. Journal of Alzheimer’s disease, 56(3):929–938, 2017.\n[67] Z. Ismail, T. K. Rajji, and K. I. Shulman. Brief cognitive screening instruments: An\nupdate. International Journal of Geriatric Psychiatry: A journal of the psychiatry\nof late life and allied sciences, 25(2):111–120, 2010.\n[68] C. R. Jack Jr, D. A. Bennett, K. Blennow, M. C. Carrillo, B. Dunn, S. B. Haeberlein,\nD. M. Holtzman, W. Jagust, F. Jessen, J. Karlawish, et al. NIAAA\nresearch\nframework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s &\nDementia, 14(4):535–562, 2018.\n[69] C. R. Jack Jr, D. S. Knopman, W. J. Jagust, R. C. Petersen, M. W. Weiner, P. S.\nAisen, L. M. Shaw, P. Vemuri, H. J. Wiste, S. D. Weigand, et al. Tracking pathophysiological processes in Alzheimer’s disease: An updated hypothetical model of\ndynamic biomarkers. The Lancet Neurology, 12(2):207–216, 2013.\n[70] C. R. Jack Jr, D. S. Knopman, W. J. Jagust, L. M. Shaw, P. S. Aisen, M. W. Weiner,\nR. C. Petersen, and J. Q. Trojanowski. Hypothetical model of dynamic biomarkers\nof the Alzheimer’s pathological cascade. The Lancet Neurology, 9(1):119–128,\n2010.\n[71] C. R. Jack Jr, P. Vemuri, H. J. Wiste, S. D. Weigand, P. S. Aisen, J. Q. Trojanowski,\nL. M. Shaw, M. A. Bernstein, R. C. Petersen, M. W. Weiner, et al. Evidence for\nordering of Alzheimer disease biomarkers. Archives of Neurology, 68(12):1526–\n1535, 2011.\n[72] J. Jacques and C. Preda. Modelbased\nclustering for multivariate functional data.\nComputational Statistics & Data Analysis, 71:92–106, 2014.\n[73] C.R.\nJiang, J. A. Aston, and J.L.\nWang. A functional approach to deconvolve\ndynamic neuroimaging data. Journal of the American Statistical Association,\n111(513):1–13, 2016.\n[74] C.R.\nJiang and L.H.\nChen. Filteringbased\napproaches for functional data classification.\nWiley Interdisciplinary Reviews: Computational Statistics, 12(4):e1490,\n2020.\n[75] M. Jo, S. Lee, Y.M.\nJeon, S. Kim, Y. Kwon, and H.J.\nKim. The role of TDP43\npropagation in neurodegenerative diseases: Integrating insights from clinical and\nexperimental studies. Experimental & Molecular Medicine, 52(10):1652–1662,\n2020.\n[76] K. A. Josephs, D. W. Dickson, N. Tosakulwong, S. D. Weigand, M. E. Murray,\nL. Petrucelli, A. M. Liesinger, M. L. Senjem, A. J. Spychalla, D. S. Knopman, et al. Rates of hippocampal atrophy and presence of postmortem\nTDP43\nin patients with\nAlzheimer’s disease: A longitudinal retrospective study. The Lancet Neurology,\n16(11):917–924, 2017.\n[77] N. Kandiah, A. Zhang, D. C. Bautista, E. Silva, S. K. S. Ting, A. Ng, and P. Assam.\nEarly detection of dementia in multilingual populations: Visual Cognitive\nAssessment Test (VCAT). Journal of Neurology, Neurosurgery & Psychiatry,\n87(2):156–160, 2016.\n[78] K. Karhunen. Über lineare methoden in der wahrscheinlichkeitsrechnung. Annales\nAcademiae Scientiarum Fennicae. Series A. 1: MathematicaPhysica,\n37:1–\n79, 1947.\n[79] M. Khanzadeh, S. Chowdhury, M. Marufuzzaman, M. A. Tschopp, and L. Bian.\nPorosity prediction: Supervisedlearning\nof thermal history for direct laser deposition.\nJournal of manufacturing systems, 47:69–82, 2018.\n[80] H. Kim and H. Kim. Functional logistic regression with fused lasso penalty. Journal\nof Statistical Computation and Simulation, 88(15):2982–2999, 2018.\n[81] D. P. Kingma and J. Ba. Adam: A method for stochastic optimization in proceedings\nof the 3rd international conference on learning representations (san diego, ca).\n2015.\n[82] W. E. Klunk, H. Engler, A. Nordberg, Y. Wang, G. Blomqvist, D. P. Holt,\nM. Bergström, I. Savitcheva, G.F.\nHuang, S. Estrada, et al. Imaging brain amyloid\nin Alzheimer’s disease with Pittsburgh CompoundB.\nAnnals of Neurology: Official\nJournal of the American Neurological Association and the Child Neurology\nSociety, 55(3):306–319, 2004.\n[83] P. Kokoszka and M. Reimherr. Introduction to Functional Data Analysis. Chapman\nand Hall/CRC, Boca Raton, 2017.\n[84] M. Krzyśko, P. Nijkamp, W. Ratajczak, and W. Wołyński. Multidimensional economic\nindicators and multivariate functional principal component analysis (MFPCA)\nin a comparative study of countries’competitiveness. Journal of Geographical\nSystems, 24:49–65, 2022.\n[85] J. K. Kueper, M. Speechley, and M. MonteroOdasso.\nThe Alzheimer’s disease\nassessment scale–cognitive subscale (ADASCog):\nModifications and responsiveness\nin predementia\npopulations. A narrative review. Journal of Alzheimer’s Disease,\n63(2):423–444, 2018.\n[86] N. M. Laird and J. H. Ware. Randomeffects\nmodels for longitudinal data. Biometrics,\n38:963–974, 1982.\n[87] K. L. Lanctôt, J. Amatniek, S. AncoliIsrael,\nS. E. Arnold, C. Ballard, J. CohenMansfield,\nZ. Ismail, C. Lyketsos, D. S. Miller, E. Musiek, et al. Neuropsychiatric\nsigns and symptoms of Alzheimer’s disease: New treatment paradigms.\nAlzheimer’s & Dementia: Translational Research & Clinical Interventions,\n3(3):440–449, 2017.\n[88] J. LanteroRodriguez,\nA. Snellman, A. L. Benedet, M. MilàAlomà,\nE. Camporesi,\nL. MontoliuGaya,\nN. J. Ashton, A. Vrillon, T. K. Karikari, J. D. Gispert, et al. Ptau235:\nA novel biomarker for staging preclinical Alzheimer’s disease. EMBO\nmolecular medicine, 13(12):e15098, 2021.\n[89] A. J. Larner. The usage of cognitive screening instruments: Test characteristics and\nsuspected diagnosis. In Cognitive Screening Instruments, pages 219–238. Springer,\nLondon, 2013.\n[90] C. Ledig, A. Schuh, R. Guerrero, R. A. Heckemann, and D. Rueckert. Structural\nbrain imaging in Alzheimer’s disease and mild cognitive impairment: Biomarker\nanalysis and shared morphometry database. Scientific reports, 8(1):1–16, 2018.\n[91] G. Lee, K. Nho, B. Kang, K.A.\nSohn, and D. Kim. Predicting Alzheimer’s disease\nprogression using multimodal\ndeep learning approach. Scientific Reports,\n9(1):1–12, 2019.\n[92] J. C. Lee, S. J. Kim, S. Hong, and Y. Kim. Diagnosis of Alzheimer’s disease\nutilizing amyloid and tau as fluid biomarkers. Experimental & Molecular Medicine,\n51(5):1–10, 2019.\n[93] X. Leng and H.G.\nMüller. Classification using functional data analysis for temporal\ngene expression data. Bioinformatics, 22(1):68–76, 2006.\n[94] A. Li, F. Li, F. Elahifasaee, M. Liu, and L. Zhang. Hippocampal shape and asymmetry\nanalysis by cascaded convolutional neural networks for Alzheimer’s disease\ndiagnosis. Brain Imaging and Behavior, 15(5):2330–2339, 2021.\n[95] B. Li and Q. Yu. Classification of functional data: A segmentation approach. Computational\nStatistics & Data Analysis, 52(10):4790–4800, 2008.\n[96] C. Li, L. Xiao, and S. Luo. Fast covariance estimation for multivariate sparse functional\ndata. Stat, 9(1):e245, 2020.\n[97] D. Li, S. Iddi, W. K. Thompson, M. C. Donohue, and A. D. N. Initiative. Bayesian\nlatent time joint mixed effect models for multicohort longitudinal data. Statistical\nMethods in Medical Research, 28(3):835–845, 2019.\n[98] H. Li, T. Pan, Y. Li, S. Chen, and G. Li. Functional principal component analysis for\nnearinfrared\nspectral data: A case study on Tricholoma matsutakeis. International\nJournal of Food Engineering, 16(8), 2020.\n[99] K. Li and S. Luo. Dynamic prediction of Alzheimer’s disease progression using\nfeatures of multiple longitudinal outcomes and timetoevent\ndata. Statistics in\nMedicine, 38(24):4804–4818, 2019.\n[100] W. Li, X. Lin, and X. Chen. Detecting Alzheimer’s disease based on 4d fMRI: An\nexploration under deep learning framework. Neurocomputing, 388:280–287, 2020.\n[101] X. Li, G. Qi, C. Yu, G. Lian, H. Zheng, S. Wu, T.F.\nYuan, and D. Zhou. Cortical\nplasticity is correlated with cognitive improvement in Alzheimer’s disease\npatients after rTMS treatment. Brain Stimulation, 14(3):503–510, 2021.\n[102] M. P. Lichtenstein, P. Carriba, R. Masgrau, A. Pujol, and E. Galea. Staging antiinflammatory\ntherapy in Alzheimer’s disease. Frontiers in Aging Neuroscience,\n2:142, 2010.\n[103] W. Liggett, L. Cazares, and O. J. Semmes. A look at mass spectral measurement.\nChance, 16(4):24–28, 2003.\n[104] N. Lin, J. Jiang, S. Guo, and M. Xiong. Functional principal component analysis\nand randomized sparse clustering algorithm for medical image analysis. PLoS One,\n10(7):e0132945, 2015.\n[105] M. Liu, D. Cheng, W. Yan, A. D. N. Initiative, et al. Classification of Alzheimer’s\ndisease by combination of convolutional and recurrent neural networks using FDGPET\nimages. Frontiers in Neuroinformatics, 12:35, 2018.\n[106] Y. Liu, L. Tan, H.F.\nWang, Y. Liu, X.K.\nHao, C.C.\nTan, T. Jiang, B. Liu, D.Q.\nZhang, and J.T.\nYu. Multiple effect of APOE genotype on clinical and neuroimaging\nbiomarkers across Alzheimer’s disease spectrum. Molecular Neurobiology,\n53(7):4539–4547, 2016.\n[107] M. Loève. Fonctions aléatoires à décomposition orthogonale exponentielle. La\nRevue Scientifique, 84:159–162, 1946.\n[108] Mayo Clinic Staff. Alzheimer’s stages: How the disease progresses.\nhttps://www.mayoclinic.org/diseases-conditions/alzheimers-disease/\nin-depth/alzheimers-stages/art-20048448. Accessed: 20211101.\n[109] M. Mehdipour Ghazi, M. Nielsen, A. Pai, M. Modat, M. Jorge Cardoso, S. Ourselin,\nand L. Sørensen. Robust parametric modeling of Alzheimer’s disease progression.\nNeuroImage, 225:117460, 2021.\n[110] S. A. Mofrad, A. J. Lundervold, A. Vik, and A. S. Lundervold. Cognitive and MRI\ntrajectories for prediction of Alzheimer’s disease. Scientific Reports, 11(1):1–10,\n2021.\n[111] R. C. Mohs, D. Knopman, R. C. Petersen, S. H. Ferris, C. Ernesto, M. Grundman,\nM. Sano, L. Bieliauskas, D. Geldmacher, C. Clark, et al. Development of cognitive\ninstruments for use in clinical trials of antidementia drugs: Additions to the\nAlzheimer’s disease assessment scale that broaden its scope. Alzheimer Disease\nand Associated Disorders, 1997.\n[112] M. Mojirsheibani and C. Shaw. Classification with incomplete functional covariates.\nStatistics & Probability Letters, 139:40–46, 2018.\n[113] A. Möller, G. Tutz, and J. Gertheiss. Random forests for functional covariates.\nJournal of Chemometrics, 30(12):715–725, 2016.\n[114] H.g.\nMüller. Functional modelling and classification of longitudinal data. Scandinavian\nJournal of Statistics, 32(2):223–240, 2005.\n[115] Z. S. Nasreddine, N. A. Phillips, V. Bédirian, S. Charbonneau, V. Whitehead,\nI. Collin, J. L. Cummings, and H. Chertkow. The Montreal Cognitive Assessment,\nMoCA: A brief screening tool for mild cognitive impairment. Journal of the American\nGeriatrics Society, 53(4):695–699, 2005.\n[116] M. Nguyen, T. He, L. An, D. C. Alexander, J. Feng, B. T. Yeo, A. D. N. Initiative,\net al. Predicting Alzheimer’s disease progression using deep recurrent neural\nnetworks. NeuroImage, 222:117203, 2020.\n[117] NIH National Institute on Aging (NIA). How biomarkers help diagnose dementia.\nhttps://www.nia.nih.gov/health/how-biomarkers-help-diagnose-dementia#\nfuture_biomarkers. Accessed: 20220201.\n[118] NIH National Institute on Aging (NIA). How is alzheimer’s disease treated? https:\n//www.nia.nih.gov/health/how-alzheimers-disease-treated. Accessed: 20220201.\n[119] M. B. T. Noor, N. Z. Zenia, M. S. Kaiser, S. A. Mamun, and M. Mahmud. Application\nof deep learning in detecting neurological disorders from magnetic resonance\nimages: A survey on the detection of Alzheimer’s disease, Parkinson’s disease\nand schizophrenia. Brain Informatics, 7(1):1–21, 2020.\n[120] T. Noori, A. R. Dehpour, A. Sureda, E. SobarzoSanchez,\nand S. Shirooie. Role\nof natural products for the treatment of Alzheimer’s disease. European Journal of\nPharmacology, 898:173974, 2021.\n[121] H.J.\nPark, K. J. Friston, C. Pae, B. Park, and A. Razi. Dynamic effective connectivity\nin resting state fMRI. NeuroImage, 180:594–608, 2018.\n[122] Penn Medicine. The 7 stages of Alzheimer’s disease. https://www.pennmedicine.\norg/updates/blogs/neuroscience-blog/2019/november/stages-of-alzheimers.\nAccessed: 20211101.\n[123] R. C. Petersen. Alzheimer’s disease: Progress in prediction. The Lancet Neurology,\n9(1):4–5, 2010.\n[124] J. Pinheiro and D. Bates. Mixedeffects\nmodels in S and SPLUS.\nSpringer, New\nYork, 2006.\n[125] J. Pinheiro, D. Bates, S. DebRoy, D. Sarkar, and R Core Team. nlme: Linear and\nNonlinear Mixed Effects Models, 2013. R package version 3.1153.\n[126] J. R. Quinlan. Induction of decision trees. Machine Learning, 1(1):81–106, 1986.\n[127] J. R. Quinlan. C4.5: Programs for machine learning. Elsevier, 2014.\n[128] G. D. Rabinovici. Controversy and progress in Alzheimer’s disease —FDA approval\nof Aducanumab. New England Journal of Medicine, 385(9):771–774, 2021.\n[129] J. Ramsay, G. Hooker, and S. Graves. Functional Data Analysis with R and MATLAB.\nSpringer, New York, 2009.\n[130] J. Ramsay and B. W. Silverman. Functional Data Analysis (2 ed.). Springer, New\nYork, 2005.\n[131] M. Ravanelli, P. Brakel, M. Omologo, and Y. Bengio. Light gated recurrent units\nfor speech recognition. IEEE Transactions on Emerging Topics in Computational\nIntelligence, 2(2):92–102, 2018.\n[132] C. Reitz. Alzheimer’s disease and the amyloid cascade hypothesis: A critical review.\nInternational journal of Alzheimer’s disease, 2012:Article ID 369808, 11\npages, 2012.\n[133] K. E. Roach, V. Pedoia, J. J. Lee, T. Popovic, T. M. Link, S. Majumdar, and R. B.\nSouza. Multivariate functional principal component analysis identifies waveform\nfeatures of gait biomechanics related to earlytomoderate\nhip osteoarthritis. Journal\nof Orthopaedic Research®, 39(8):1722–1731, 2021.\n[134] F. Rossi and N. Villa. Support vector machine for functional data classification.\nNeurocomputing, 69(79):\n730–742, 2006.\n[135] I. Saied, T. Arslan, and S. Chandran. Classification of Alzheimer’s disease using\nRF signals and machine learning. IEEE Journal of Electromagnetics, RF and\nMicrowaves in Medicine and Biology, 6(1), 2022.\n[136] A. Sarica, R. Vasta, F. Novellino, M. G. Vaccaro, A. Cerasa, A. Quattrone, A. D. N.\nInitiative, et al. MRI asymmetry index of hippocampal subfields increases through\nthe continuum from the mild cognitive impairment to the Alzheimer’s disease.\nFrontiers in Neuroscience, page 576, 2018.\n[137] S. W. Scheff, D. A. Price, F. A. Schmitt, M. A. Scheff, and E. J. Mufson. Synaptic\nloss in the inferior temporal gyrus in mild cognitive impairment and alzheimer’s\ndisease. Journal of Alzheimer’s Disease, 24(3):547–557, 2011.\n[138] P. Scheltens, D. Leys, F. Barkhof, D. Huglo, H. Weinstein, P. Vermersch, M. Kuiper,\nM. Steinling, E. C. Wolters, and J. Valk. Atrophy of medial temporal lobes on\nMRI in ” probable” Alzheimer’s disease and normal ageing: Diagnostic value and\nneuropsychological correlates. Journal of Neurology, Neurosurgery & Psychiatry,\n55(10):967–972, 1992.\n[139] S. A. Sikkes, E. S. de Langede\nKlerk, Y. A. Pijnenburg, F. Gillissen, R. Romkes,\nD. L. Knol, B. M. Uitdehaag, and P. Scheltens. A new informantbased\nquestionnaire for instrumental activities of daily living in dementia. Alzheimer’s & Dementia,\n8(6):536–543, 2012.\n[140] A. Singleton, M. Farrer, J. Johnson, A. Singleton, S. Hague, J. Kachergus, M. Hulihan,\nT. Peuralinna, A. N. Dutra, S. Lincoln, et al. αsynuclein\nlocus triplication\ncauses Parkinson’s disease. Science, 302(5646):841–842, 2003.\n[141] R. Smith, T. Mukerji, and T. Lupo. Correlating geologic and seismic data with\nunconventional resource production curves using machine learning. Geophysics,\n84(2):O39–O47, 2019.\n[142] T. A. Snijders and R. J. Bosker. Multilevel analysis: An introduction to basic and\nadvanced multilevel modeling (2 ed.). Sage Publications, London, 2011.\n[143] H. Sørensen, J. Goldsmith, and L. M. Sangalli. An introduction with medical applications\nto functional data analysis. Statistics in Medicine, 32(30):5222–5240,\n2013.\n[144] R. A. Sperling, P. S. Aisen, L. A. Beckett, D. A. Bennett, S. Craft, A. M. Fagan,\nT. Iwatsubo, C. R. Jack Jr, J. Kaye, T. J. Montine, et al. Toward defining\nthe preclinical stages of Alzheimer’s disease: Recommendations from the National\nInstitute on AgingAlzheimer’s\nAssociation workgroups on diagnostic guidelines\nfor Alzheimer’s disease. Alzheimer’s & dementia, 7(3):280–292, 2011.\n[145] S. Srivastava, R. Ahmad, and S. K. Khare. Alzheimer’s disease and its treatment\nby different approaches: A review. European Journal of Medicinal Chemistry,\n216:113320, 2021.\n[146] J. E. Storey, J. T. Rowland, D. A. Conforti, and H. G. Dickson. The Rowland universal\ndementia assessment scale (RUDAS): A multicultural cognitive assessment\nscale. International Psychogeriatrics, 16(1):13–31, 2004.\n[147] Y. Su and C.C.\nJ. Kuo. On extended long shortterm\nmemory and dependent bidirectional\nrecurrent neural network. Neurocomputing, 356:151–161, 2019.\n[148] Taiwan Alzheimer Disease Association. 認識失智症. http://www.tada2002.org.\ntw/About/IsntDementia, 04 2021. Accessed: 20210928.\n[149] M. Tanveer, B. Richhariya, R. Khan, A. Rashid, P. Khanna, M. Prasad, and C. Lin.\nMachine learning techniques for the diagnosis of Alzheimer’s disease: A review.\nACM Transactions on Multimedia Computing, Communications, and Applications\n(TOMM), 16(1s):1–35, 2020.\n[150] S. J. Teipel, W. Bayer, G. E. Alexander, Y. Zebuhr, D. Teichberg, L. Kulic, M. B.\nSchapiro, H.J.\nMöller, S. I. Rapoport, and H. Hampel. Progression of Corpus\nCallosum Atrophy in Alzheimer Disease. Archives of Neurology, 59(2):243–248,\n02 2002.\n[151] C. G. Thomas, R. A. Harshman, and R. S. Menon. Noise reduction in BOLDbased\nfMRI using component analysis. Neuroimage, 17(3):1521–1537, 2002.\n[152] M. Torso, M. Bozzali, G. Zamboni, M. Jenkinson, S. A. Chance, and A. D. N.\nInitiative. Detection of Alzheimer’s disease using cortical diffusion tensor imaging.\nHuman Brain Mapping, 42(4):967–977, 2021.\n[153] D. Tosun, Z. Demir, D. P. Veitch, D. Weintraub, P. Aisen, C. R. Jack Jr,\nW. J. Jagust, R. C. Petersen, A. J. Saykin, L. M. Shaw, et al. Contribution of\nAlzheimer’s biomarkers and risk factors to cognitive impairment and decline across\nthe Alzheimer’s disease continuum. Alzheimer’s & Dementia, 2021.\n[154] G. Van Rossum and F. L. Drake Jr. Python reference manual. Centrum voor\nWiskunde en Informatica Amsterdam, 1995.\n[155] M. Vernooij, F. Pizzini, R. Schmidt, M. Smits, T. Yousry, N. Bargallo, G. Frisoni,\nS. Haller, and F. Barkhof. Dementia imaging in clinical practice: A europeanwide\nsurvey of 193 centres and conclusions by the ESNR working group. Neuroradiology,\n61(6):633–642, 2019.\n[156] R. Viviani, G. Grön, and M. Spitzer. Functional principal component analysis of\nfMRI data. Human brain mapping, 24(2):109–129, 2005.\n[157] M. Walterfang, E. Luders, J. C. Looi, P. Rajagopalan, D. Velakoulis, P. M. Thompson,\nO. Lindberg, P. Östberg, L. E. Nordin, L. Svensson, et al. Shape analysis of\nthe corpus callosum in Alzheimer’s disease and frontotemporal lobar degeneration\nsubtypes. Journal of Alzheimer’s Disease, 40(4):897–906, 2014.\n[158] J.L.\nWang, J.M.\nChiou, and H.G.\nMüller. Functional data analysis. Annual Review\nof Statistics and Its Application, 3:257–295, 2016.\n[159] L. Wang, Y. Zang, Y. He, M. Liang, X. Zhang, L. Tian, T. Wu, T. Jiang, and K. Li.\nChanges in hippocampal connectivity in the early stages of Alzheimer’s disease:\nEvidence from resting state fMRI. Neuroimage, 31(2):496–504, 2006.\n[160] Y. Wei, G. Xiao, H. Deng, H. Chen, M. Tong, G. Zhao, and Q. Liu. Hyperspectral\nimage classification using FPCAbased\nkernel extreme learning machine. Optik,\n126(23):3942–3948, 2015.\n[161] R. K. Wong, Y. Li, and Z. Zhu. Partially linear functional additive models for\nmultivariate functional data. Journal of the American Statistical Association,\n114(525):406–418, 2019.\n[162] World Health Organization. Dementia. https://www.who.int/news-room/\nfact-sheets/detail/dementia. Accessed: 2021-09-28.\n[163] World Health Organization. Dementia: a public health priority. https://www.who.\nint/publications/i/item/dementia-a-public-health-priority. Accessed: 2021-09-28.\n[164] World Health Organization. The top 10 causes of death. https://www.who.int/\nnews-room/fact-sheets/detail/the-top-10-causes-of-death. Accessed: 2021-09-28.\n[165] Y. Wu and Y. Liu. Functional robust support vector machines for sparse and\nirregular longitudinal data. Journal of computational and Graphical Statistics,\n22(2):379–395, 2013.\n[166] S. Xie. Wavelet power spectral domain functional principal component analysis for\nfeature extraction of epileptic EEGs. Computation, 9(7):78, 2021.\n[167] F. Xue, F. Tan, Z. Ye, J. Chen, and Y. Wei. Spectralspatial\nclassification of hyperspectral\nimage using improved functional principal component analysis. IEEE\nGeoscience and Remote Sensing Letters, 19:1–5, 2021.\n[168] B. Yang, H. Yu, M. Xing, R. He, R. Liang, and L. Zhou. The relationship between\ncognition and depressive symptoms, and factors modifying this association,\nin Alzheimer’s disease: A multivariate multilevel model. Archives of Gerontology\nand Geriatrics, 72:25–31, 2017.\n[169] L. Yang, J. Yan, X. Jin, Y. Jin, W. Yu, S. Xu, and H. Wu. Screening for dementia\nin older adults: Comparison of MiniMental\nState Examination, MiniCog,\nClock\nDrawing Test and AD8. PLOS ONE, 11(12):1–9, 12 2016.\n[170] F. Yao, E. Lei, and Y. Wu. Effective dimension reduction for sparse functional data.\nBiometrika, 102(2):421–437, 2015.\n[171] F. Yao, H.G.\nMüller, and J.L.\nWang. Functional data analysis for sparse longitudinal\ndata. Journal of the American statistical association, 100(470):577–590,\n2005.\n[172] F. Yao, Y. Wu, and J. Zou. Probabilityenhanced\neffective dimension reduction for\nclassifying sparse functional data. Test, 25(1):1–22, 2016.\n[173] L. Zhang, M. Wang, M. Liu, and D. Zhang. A survey on deep learning for\nneuroimagingbased\nbrain disorder analysis. Frontiers in Neuroscience, page 779,\n2020.\n[174] 台灣神經學學會Taiwan Neurological Society. 台灣神經學學會會訊2020 年\n01 月第80 期. http://www.neuro.org.tw/files/newsletter/080.pdf. Accessed:\n2021-09-28.\n[175] 衛生福利部Ministry of Health and Welfare. 失智症防治照護政策綱\n領暨行動方案2.0(含工作項目)(2021 年版). https://1966.gov.tw/LTC/\ncp-4020-42469-201.html. Accessed: 2021-09-28.\n[176] 衛生福利部中央健康保險署National Health Insurance Administration, Ministry\nof Health and Welfare. 最新版藥品給付規定內容\n第1 節神經系統藥\n物drugs acting on the nervous system. https://www.nhi.gov.tw/Content_List.\naspx?n=E70D4F1BD029DC37&topn=5FE8C9FEAE863B46. Update: 20220224,\nAccessed: 2022-03-02.\n[177] 衛生福利部統計處Department of Statistics, Ministry of Health and Welfare.\n國際失智症日衛生福利統計通報. https://www.mohw.gov.tw/\ndl-71799-1d824fee-a486-4504-9c7d-5d819c6848b2.html. Accessed: 2021-09-28. 描述 博士
國立政治大學
統計學系
99354501資料來源 http://thesis.lib.nccu.edu.tw/record/#G0099354501 資料類型 thesis dc.contributor.advisor 劉惠美 zh_TW dc.contributor.advisor Liu, Hui-Mei en_US dc.contributor.author (Authors) 李詠玄 zh_TW dc.contributor.author (Authors) Lee, Yong-Shiuan en_US dc.creator (作者) 李詠玄 zh_TW dc.creator (作者) Lee, Yong-Shiuan en_US dc.date (日期) 2022 en_US dc.date.accessioned 1-Aug-2022 17:13:51 (UTC+8) - dc.date.available 1-Aug-2022 17:13:51 (UTC+8) - dc.date.issued (上傳時間) 1-Aug-2022 17:13:51 (UTC+8) - dc.identifier (Other Identifiers) G0099354501 en_US dc.identifier.uri (URI) https://ah.lib.nccu.edu.tw/item?item_id=160360 - dc.description (描述) 博士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 統計學系 zh_TW dc.description (描述) 99354501 zh_TW dc.description.abstract (摘要) 自二十世紀晚期對於探討阿茲海默症成因、病情發展與有效治療方式的研究大量增加。其中最重要的目標之一即為於早期診斷出阿茲海默症，也就是輕度認知障礙。診斷輕度認知障礙或阿茲海默症即是統計上的分類問題。通常阿茲海默症的相關研究資料皆為長期追蹤資料，由於資料收集的方式，使得資料多為稀疏性且不規則間隔的資料。再者，基於近年來醫學診斷工具，尤其是腦部顯影的技術進步與普及，阿茲海默症資料更常為具有高維度的資料。傳統上常用的統計分類方法對於此類資料型態有其侷限性。本研究首先將資料的變數視為只具有少數觀察值的函數，使用函數主成分分析工具來重建高維度、稀疏且不規則間隔的資料，使資料收集區間內的所有觀察時間點皆能有函數的估計值。接續再利用遞迴類神經網路中專門針對時間序列資料的長短期記憶類神經網路，來對研究對象做診斷的分類。本研究的實證結果指出在最佳情境下，此作法使用較多觀察值於訓練資料集，以及使用較多的輸入變數，能夠正確辨認出最多的早期輕度認知障礙者（十一個患者中正確辨認出五個）。顯示此法對於辨認早期的輕度認知障礙有較大的潛力。針對阿茲海默症此類醫學研究中常見的不平衡資料，未來可考慮加入重新採樣的方法或是成本考量的分類方法進一步發展優化本文所提出之程序。 zh_TW dc.description.abstract (摘要) Since the late 20th century, researches of Alzheimer’s disease intending to better understand the causes, the progression, and effective treatments of this disease have boosted. One of the most important purposes of these researches is to detect the disease at early stages, that is, the diagnosis of mild cognitive impairment. The diagnosis is certainly the classification problem in statistics. The research data of Alzheimer’s disease are usually longitudinal, which can be very sparse and irregularlyspaced as a result of data collection process. Additionally, the research data can also have high imensional features due to improvement in clinical neuroimaging techniques. Classical approaches for classification have limitations in using the sparse and irregular, highdimensional, longitudinal data. This study is the first to implement the tool of the functional principal component analysis to reconstruct the whole\nfunctions of all variables during the period, and then to apply the long shortterm memory networks, a recurrent neural network designed for time series data, for classification. The empirical results show that in the bestcase scenario this method identifies 5 out of 11 MCI cases in the testing dataset while the other methods only accurately predict 0 or 1 MCI case. The results suggest that this procedure has great potential for early detection of Alzheimer’s disease. The proposed method can further be developed for imbalanced data with resampling or costsensitive classification techniques. en_US dc.description.tableofcontents 誌謝 i\n摘要 ii\nAbstract iv\nContents vi\nList of Figures viii\nList of Tables xi\n1 Introduction 1\n1.1 Dementia 2\n1.2 Alzheimer’s Disease 6\n1.3 Diagnosis of Alzheimer’s Disease 7\n1.4 Treatment for Alzheimer’s disease 13\n1.5 Outline of The Dissertation 14\n2 Related Work 15\n2.1 The Mixedeffects Model 16\n2.2 Linear Discriminant Analysis and Related Methods 18\n2.3 Functional Principal Component Analysis 20\n2.3.1 Univariate FPCA 21\n2.3.2 Multivariate FPCA 24\n2.3.3 Application of FPCA to Medical data 26\n2.4 Classification for Longitudinal Data 27\n2.4.1 Functional Data Classification 27\n2.4.2 Deep Learning Models 29\n3 Analysis 31\n3.1 Data Description 31\n3.2 Multilevel Modeling for The Scores of MMSE and ADASCog13\n35\n3.2.1 TwoLevel Mixed Effects Model With Demographic and Neuroimaging Variables 35\n3.2.2 TwoLevel Mixed Effects Model Including Fixed Effect of Grouping by Final Status 39\n3.3 Univariate Functional PCA of Midanterior\nCorpus Callosum 41\n3.4 Multivariate Functional PCA 49\n3.4.1 Multivariate Functional PCA of Left and Right Hippocampus Volumes 49\n3.4.2 Multivariate Functional PCA of Cognitive Assessment Scales and fMRI Variables 56\n3.5 Classification 58\n3.5.1 Classification by Flexible Discriminant Analysis and Regularized Discriminant Analysis 59\n3.5.2 Classification by LSTM Using Features Reconstructed From FPCA 63\n4 Discussion and Conclusions 69\nReference 71\nAppendix A: Tables of the Twolevel Growth Models 94 zh_TW dc.format.extent 4503222 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0099354501 en_US dc.subject (關鍵詞) 阿茲海默症 zh_TW dc.subject (關鍵詞) 函數主成分分析 zh_TW dc.subject (關鍵詞) 遞迴類神經網路 zh_TW dc.subject (關鍵詞) 長短期記憶類神經網路 zh_TW dc.subject (關鍵詞) 長期追蹤資料 zh_TW dc.subject (關鍵詞) Alzheimer’s disease en_US dc.subject (關鍵詞) Functional principal component analysis en_US dc.subject (關鍵詞) Recurrent neural networks en_US dc.subject (關鍵詞) Long short-term memory networks en_US dc.subject (關鍵詞) Longitudinal data en_US dc.title (題名) 運用函數主成分分析於阿茲海默症之診斷 zh_TW dc.title (題名) Application of functional principal component analysis to diagnosis of Alzheimer’s disease en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) [1] M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat,\nG. Irving, M. Isard, et al. Tensorflow: A system for largescale\nmachine learning.\nIn 12th {USENIX} Symposium on Operating Systems Design and Implementation\n({OSDI} 16), pages 265–283, 2016.\n[2] A. Anoop, P. K. Singh, R. S. Jacob, and S. K. Maji. CSF biomarkers for Alzheimer’s\ndisease diagnosis. International journal of Alzheimer’s disease, 2010:Article ID\n606802, 12 pages, 2010.\n[3] A. Association. 2020 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia,\n16(3):391–460, 2020.\n[4] A. Association. 2021 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia,\n17(3):327–406, 2021.\n[5] S. Balakrishnan and D. Madigan. Decision trees for functional variables. In Sixth\nInternational Conference on Data Mining (ICDM’06), pages 798–802. IEEE, 2006.\n[6] E. Belli and S. Vantini. Measure inducing classification and regression trees for\nfunctional data. Statistical Analysis and Data Mining: The ASA Data Science Journal,\n2021.\n[7] Y. Bengio. Learning deep architectures for AI. Foundations and Trends in Signal\nProcessing, 2(1):1–127, 2009.\n[8] J. R. Berrendero, A. Justel, and M. Svarc. Principal components for multivariate\nfunctional data. Computational Statistics & Data Analysis, 55(9):2619–2634, 2011.\n[9] M. Bertoux, J. Lagarde, F. Corlier, L. Hamelin, J.F.\nMangin, O. Colliot, M. Chupin,\nM. N. Braskie, P. M. Thompson, M. Bottlaender, et al. Sulcal morphology in\nAlzheimer’s disease: An effective marker of diagnosis and cognition. Neurobiology\nof Aging, 84:41–49, 2019.\n[10] M. C. Biagioni and J. E. Galvin. Using biomarkers to improve detection of\nAlzheimer’s disease. Neurodegenerative Disease Management, 1(2):127–139,\n2011.\n[11] S. Borson, J. Scanlan, M. Brush, P. Vitaliano, and A. Dokmak. The MiniCog:\nA\ncognitive ‘vital signs’measure for dementia screening in multilingual\nelderly.\nInternational journal of geriatric psychiatry, 15(11):1021–1027, 2000.\n[12] S. Borson, J. M. Scanlan, P. Chen, and M. Ganguli. The MiniCog\nas a screen\nfor dementia: Validation in a populationbased\nsample. Journal of the American\nGeriatrics Society, 51(10):1451–1454, 2003.\n[13] L. Breiman. Bagging predictors. Machine Learning, 24(2):123–140, 1996.\n[14] L. Breiman. Random forests. Machine Learning, 45(1):5–32, 2001.\n[15] L. Breiman, J. H. Friedman, R. A. Olshen, and C. J. Stone. Classification and\nRegression Trees. Chapman & Hall/CRC, New York., 1984.\n[16] A. M. Brickman, J. J. Manly, L. S. Honig, D. Sanchez, D. ReyesDumeyer,\nR. A.\nLantigua, P. J. Lao, Y. Stern, J. P. Vonsattel, A. F. Teich, et al. Plasma ptau181,ptau217,\nand other bloodbased\nAlzheimer’s disease biomarkers in a multiethnic,\ncommunity study. Alzheimer’s & Dementia, 17(8):1353–1364, 2021.\n[17] R. S. Bucks, D. Ashworth, G. Wilcock, and K. Siegfried. Assessment of activities\nof daily living in dementia: Development of the bristol activities of daily living\nscale. Age and ageing, 25(2):113–120, 1996.\n[18] H. Buschke, G. Kuslansky, M. Katz, W. F. Stewart, M. J. Sliwinski, H. M. Eckholdt,\nand R. B. Lipton. Screening for dementia with the memory impairment screen.\nNeurology, 52(2):231–231, 1999.\n[19] B. D. Carpenter, C. Xiong, E. K. Porensky, M. M. Lee, P. J. Brown, M. Coats,\nD. Johnson, and J. C. Morris. Reaction to a dementia diagnosis in individuals with\nAlzheimer’s disease and mild cognitive impairment. Journal of the American Geriatrics\nSociety, 56(3):405–412, 2008.\n[20] L.H.\nChen and C.R.\nJiang. Multidimensional\nfunctional principal component\nanalysis. Statistics and Computing, 27(5):1181–1192, 2017.\n[21] W.C.\nCheng, L.H.\nChen, C.R.\nJiang, Y.M.\nDeng, D.W.\nWang, C.H.\nLin, R. Jou,\nJ.K.\nWang, and Y.L.\nWang. Sensible functional linear discriminant analysis effectively\ndiscriminates enhanced Raman spectra of Mycobacterium species. Analytical\nChemistry, 93(5):2785–2792, 2021. PMID: 33480698.\n[22] R. Chin, A. Ng, K. Narasimhalu, and N. Kandiah. Utility of the AD8 as a selfrating\ntool for cognitive impairment in an Asian population. American Journal of\nAlzheimer’s Disease & Other Dementias®, 28(3):284–288, 2013.\n[23] J.M.\nChiou, Y.T.\nChen, and Y.F.\nYang. Multivariate functional principal component\nanalysis: A normalization approach. Statistica Sinica, pages 1571–1596,\n2014.\n[24] K. Cho, B. Van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk,\nand Y. Bengio. Learning phrase representations using RNN encoderdecoder\nfor\nstatistical machine translation. In Proceedings of the 2014 Conference on Empirical\nMethods in Natural Language Processing (EMNLP), page 1724–1734. Association\nfor Computational Linguistics (ACL), Oct. 2014.\n[25] S. H. Cho, S. Woo, C. Kim, H. J. Kim, H. Jang, B. C. Kim, S. E. Kim, S. J. Kim, J. P.\nKim, Y. H. Jung, et al. Disease progression modelling from preclinical Alzheimer’\ns disease (AD) to AD dementia. Scientific reports, 11(1):1–10, 2021.\n[26] F. Chollet et al. Keras. urlhttps://github.com/fchollet/keras, 2015.\n[27] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio. Empirical evaluation of gated recurrent\nneural networks on sequence modeling. arXiv preprint arXiv:1412.3555,\n2014.\n[28] M. Conceição, A. KroneMartins,\nand A. da Silva. FPCA emulation of cosmological\nsimulations. In 2021 IEEE 17th International Conference on eScience\n(eScience), pages 225–226. IEEE, 2021.\n[29] C. Cortes and V. Vapnik. Support vector machine. Machine Learning, 20(3):273–\n297, 1995.\n[30] R. Cui, M. Liu, A. D. N. Initiative, et al. RNNbased\nlongitudinal analysis for\ndiagnosis of Alzheimer’s disease. Computerized Medical Imaging and Graphics,\n73:1–10, 2019.\n[31] J. M. Cuttler, E. Abdellah, Y. Goldberg, S. AlShamaa,\nS. P. Symons, S. E. Black,\nand M. Freedman. Low doses of ionizing radiation as a treatment for Alzheimer’\ns disease: A pilot study. Journal of Alzheimer’s Disease, 80(3):1119–1128, 2021.\n[32] A. Delaigle and P. Hall. Achieving near perfect classification for functional\ndata. Journal of the Royal Statistical Society: Series B (Statistical Methodology),\n74(2):267–286, 2012.\n[33] A. Delaigle and P. Hall. Classification using censored functional data. Journal of\nthe American Statistical Association, 108(504):1269–1283, 2013.\n[34] A. Delaigle, P. Hall, and N. Bathia. Componentwise classification and clustering\nof functional data. Biometrika, 99(2):299–313, 2012.\n[35] L. Deng and D. Yu. Deep learning: Methods and applications. Foundations and\nTrends in Signal Processing, 7(3–4):197–387, 2014.\n[36] B. Dunn, P. Stein, and P. Cavazzoni. Approval of Aducanumab for Alzheimer\ndisease—The FDA’s perspective. JAMA Internal Medicine, 181(10):1276–1278,\n2021.\n[37] S. ElSappagh,\nT. Abuhmed, S. R. Islam, and K. S. Kwak. Multimodal multitask\ndeep learning model for Alzheimer’s disease progression detection based on time\nseries data. Neurocomputing, 412:197–215, 2020.\n[38] A. Ezzati, M. J. Katz, A. R. Zammit, M. L. Lipton, M. E. Zimmerman, M. J. Sliwinski,\nand R. B. Lipton. Differential association of left and right hippocampal\nvolumes with verbal episodic and spatial memory in older adults. Neuropsychologia,\n93:380–385, 2016.\n[39] J. Fan and I. Gijbels. Local Polynomial Modelling and Its Applications. Chapman\n& Hall/CRC, London, 1996.\n[40] C. Feng, A. Elazab, P. Yang, T. Wang, F. Zhou, H. Hu, X. Xiao, and B. Lei. Deep\nlearning framework forAlzheimer’s disease diagnosis via 3DCNN\nand FSBiLSTM.\nIEEE Access, 7:63605–63618, 2019.\n[41] A. Field, J. Miles, and Z. Field. Discovering statistics using R. Sage Publications,\n2012.\n[42] M. F. Folstein, S. E. Folstein, and P. R. McHugh. “Minimental\nstate”: A practical\nmethod for grading the cognitive state of patients for the clinician. Journal of\npsychiatric research, 12(3):189–198, 1975.\n[43] P. Forouzannezhad, A. Abbaspour, C. Fang, M. Cabrerizo, D. Loewenstein,\nR. Duara, and M. Adjouadi. A survey on applications and analysis methods of\nfunctional magnetic resonance imaging for Alzheimer’s disease. Journal of neuroscience\nmethods, 317:121–140, 2019.\n[44] S. Förster, B. H. Yousefi, H.J.\nWester, E. Klupp, A. Rominger, H. Förstl, A. Kurz,\nT. Grimmer, and A. Drzezga. Quantitative longitudinal interrelationships between\nbrain metabolism and amyloid deposition during a 2year\nfollowup\nin patients with\nearly Alzheimer’s disease. European journal of nuclear medicine and molecular\nimaging, 39(12):1927–1936, 2012.\n[45] J. H. Friedman. Regularized discriminant analysis. Journal of the American Statistical\nAssociation, 84(405):165–175, 1989.\n[46] A. Gajardo, C. Carroll, Y. Chen, X. Dai, J. Fan, P. Z. Hadjipantelis, K. Han, H. Ji,\nH.G.\nMüller, and J.L.\nWang. fdapace: Functional Data Analysis and Empirical\nDynamics, 2021. R package version 0.5.7.\n[47] T. P. Garcia and K. Marder. Statistical approaches to longitudinal data analysis in\nneurodegenerative diseases: Huntington’s disease as a model. Current Neurology\nand Neuroscience Reports, 17(2):1–9, 2017.\n[48] S. Gauthier, P. RosaNeto,\nJ. A. Morais, C. Webster, et al. World Alzheimer report\n2021 Journey\nthrough the diagnosis of dementia. https://www.alzint.org/\nresource/world-alzheimer-report-2021/. Accessed: 20210928.\n[49] I. Gélinas, L. Gauthier, M. McIntyre, and S. Gauthier. Development of a functional\nmeasure for persons with Alzheimer’s disease: the disability assessment\nfor dementia. American Journal of Occupational Therapy, 53(5):471–481, 1999.\n[50] M. M. Ghazi, M. Nielsen, A. Pai, M. J. Cardoso, M. Modat, S. Ourselin,\nL. Sørensen, A. D. N. Initiative, et al. Training recurrent neural networks robust\nto incomplete data: Application to Alzheimer’s disease progression modeling.\nMedical Image Analysis, 53:39–46, 2019.\n[51] Y. Gupta, R. K. Lama, G.R.\nKwon, M. W. Weiner, P. Aisen, M. Weiner, R. Petersen,\nC. R. Jack Jr, W. Jagust, J. Q. Trojanowki, et al. Prediction and classification\nof Alzheimer’s disease based on combined features from apolipoproteinE\ngenotype,\ncerebrospinal fluid, MR, and FDGPET\nimaging biomarkers. Frontiers in\nComputational Neuroscience, 13:72, 2019.\n[52] Y. Gupta, K. H. Lee, K. Y. Choi, J. J. Lee, B. C. Kim, G. R. Kwon, N. R. C. for\nDementia, and A. D. N. Initiative. Early diagnosis of Alzheimer’s disease using\ncombined features from voxelbased\nmorphometry and cortical, subcortical, and\nhippocampus regions of MRI T1 brain images. PLoS One, 14(10):e0222446, 2019.\n[53] C. Happ and S. Greven. Multivariate functional principal component analysis for\ndata observed on different (dimensional) domains. Journal of the American Statistical\nAssociation, 113(522):649–659, 2018.\n[54] C. HappKurz.\nObjectoriented\nsoftware for functional data. Journal of Statistical\nSoftware, 93(5):1–38, 2020.\n[55] C. HappKurz.\nMFPCA: Multivariate Functional Principal Component Analysis\nfor Data Observed on Different Dimensional Domains, 2021. R package version\n1.39.\n[56] J. A. Hardy and G. A. Higgins. Alzheimer’s disease: The amyloid cascade hypothesis.\nScience, 256(5054):184–186, 1992.\n[57] K. Hasenstab, A. Scheffler, D. Telesca, C. A. Sugar, S. Jeste, C. DiStefano, and\nD. Şentürk. A multidimensional\nfunctional principal components analysis of EEG\ndata. Biometrics, 73(3):999–1009, 2017.\n[58] T. Hastie. [Flexible Parsimonious Smoothing and Additive Modeling]: Discussion.\nTechnometrics, 31(1):23–29, 1989.\n[59] T. Hastie, A. Buja, and R. Tibshirani. Penalized discriminant analysis. The Annals\nof Statistics, 23(1):73–102, 1995.\n[60] T. Hastie, R. Tibshirani, and A. Buja. Flexible discriminant analysis by optimal\nscoring. Journal of the American Statistical Association, 89(428):1255–1270,\n1994.\n[61] S. Hochreiter and J. Schmidhuber. Long shortterm\nmemory. Neural Computation,\n9(8):1735–1780, 1997.\n[62] H. Hodkinson. Evaluation of a mental test score for assessment of mental impairment\nin the elderly. Age and ageing, 1(4):233–238, 1972.\n[63] W. Huang, Y. Zhou, L. Tu, Z. Ba, J. Huang, N. Huang, and Y. Luo. TDP43:\nFrom\nAlzheimer’s disease to limbicpredominant\nagerelated\nTDP43\nencephalopathy.\nFrontiers in Molecular Neuroscience, 13:26, 2020.\n[64] S. Iddi, D. Li, P. S. Aisen, M. S. Rafii, W. K. Thompson, and M. C. Donohue.\nPredicting the course of Alzheimer’s progression. Brain Informatics, 6(1):1–18,\n2019.\n[65] S. Ioffe and C. Szegedy. Batch normalization: Accelerating deep network training\nby reducing internal covariate shift. In International Conference on Machine\nLearning, pages 448–456. PMLR, 2015.\n[66] Z. Ismail, L. AgüeraOrtiz,\nH. Brodaty, A. Cieslak, J. Cummings, C. E. Fischer,\nS. Gauthier, Y. E. Geda, N. Herrmann, J. Kanji, et al. The Mild Behavioral\nImpairment Checklist (MBIC):\nA rating scale for neuropsychiatric symptoms in\npredementia\npopulations. Journal of Alzheimer’s disease, 56(3):929–938, 2017.\n[67] Z. Ismail, T. K. Rajji, and K. I. Shulman. Brief cognitive screening instruments: An\nupdate. International Journal of Geriatric Psychiatry: A journal of the psychiatry\nof late life and allied sciences, 25(2):111–120, 2010.\n[68] C. R. Jack Jr, D. A. Bennett, K. Blennow, M. C. Carrillo, B. Dunn, S. B. Haeberlein,\nD. M. Holtzman, W. Jagust, F. Jessen, J. Karlawish, et al. NIAAA\nresearch\nframework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s &\nDementia, 14(4):535–562, 2018.\n[69] C. R. Jack Jr, D. S. Knopman, W. J. Jagust, R. C. Petersen, M. W. Weiner, P. S.\nAisen, L. M. Shaw, P. Vemuri, H. J. Wiste, S. D. Weigand, et al. Tracking pathophysiological processes in Alzheimer’s disease: An updated hypothetical model of\ndynamic biomarkers. The Lancet Neurology, 12(2):207–216, 2013.\n[70] C. R. Jack Jr, D. S. Knopman, W. J. Jagust, L. M. Shaw, P. S. Aisen, M. W. Weiner,\nR. C. Petersen, and J. Q. Trojanowski. Hypothetical model of dynamic biomarkers\nof the Alzheimer’s pathological cascade. The Lancet Neurology, 9(1):119–128,\n2010.\n[71] C. R. Jack Jr, P. Vemuri, H. J. Wiste, S. D. Weigand, P. S. Aisen, J. Q. Trojanowski,\nL. M. Shaw, M. A. Bernstein, R. C. Petersen, M. W. Weiner, et al. Evidence for\nordering of Alzheimer disease biomarkers. Archives of Neurology, 68(12):1526–\n1535, 2011.\n[72] J. Jacques and C. Preda. Modelbased\nclustering for multivariate functional data.\nComputational Statistics & Data Analysis, 71:92–106, 2014.\n[73] C.R.\nJiang, J. A. Aston, and J.L.\nWang. A functional approach to deconvolve\ndynamic neuroimaging data. Journal of the American Statistical Association,\n111(513):1–13, 2016.\n[74] C.R.\nJiang and L.H.\nChen. Filteringbased\napproaches for functional data classification.\nWiley Interdisciplinary Reviews: Computational Statistics, 12(4):e1490,\n2020.\n[75] M. Jo, S. Lee, Y.M.\nJeon, S. Kim, Y. Kwon, and H.J.\nKim. The role of TDP43\npropagation in neurodegenerative diseases: Integrating insights from clinical and\nexperimental studies. Experimental & Molecular Medicine, 52(10):1652–1662,\n2020.\n[76] K. A. Josephs, D. W. Dickson, N. Tosakulwong, S. D. Weigand, M. E. Murray,\nL. Petrucelli, A. M. Liesinger, M. L. Senjem, A. J. Spychalla, D. S. Knopman, et al. Rates of hippocampal atrophy and presence of postmortem\nTDP43\nin patients with\nAlzheimer’s disease: A longitudinal retrospective study. The Lancet Neurology,\n16(11):917–924, 2017.\n[77] N. Kandiah, A. Zhang, D. C. Bautista, E. Silva, S. K. S. Ting, A. Ng, and P. Assam.\nEarly detection of dementia in multilingual populations: Visual Cognitive\nAssessment Test (VCAT). Journal of Neurology, Neurosurgery & Psychiatry,\n87(2):156–160, 2016.\n[78] K. Karhunen. Über lineare methoden in der wahrscheinlichkeitsrechnung. Annales\nAcademiae Scientiarum Fennicae. Series A. 1: MathematicaPhysica,\n37:1–\n79, 1947.\n[79] M. Khanzadeh, S. Chowdhury, M. Marufuzzaman, M. A. Tschopp, and L. Bian.\nPorosity prediction: Supervisedlearning\nof thermal history for direct laser deposition.\nJournal of manufacturing systems, 47:69–82, 2018.\n[80] H. Kim and H. Kim. Functional logistic regression with fused lasso penalty. Journal\nof Statistical Computation and Simulation, 88(15):2982–2999, 2018.\n[81] D. P. Kingma and J. Ba. Adam: A method for stochastic optimization in proceedings\nof the 3rd international conference on learning representations (san diego, ca).\n2015.\n[82] W. E. Klunk, H. Engler, A. Nordberg, Y. Wang, G. Blomqvist, D. P. Holt,\nM. Bergström, I. Savitcheva, G.F.\nHuang, S. Estrada, et al. Imaging brain amyloid\nin Alzheimer’s disease with Pittsburgh CompoundB.\nAnnals of Neurology: Official\nJournal of the American Neurological Association and the Child Neurology\nSociety, 55(3):306–319, 2004.\n[83] P. Kokoszka and M. Reimherr. Introduction to Functional Data Analysis. Chapman\nand Hall/CRC, Boca Raton, 2017.\n[84] M. Krzyśko, P. Nijkamp, W. Ratajczak, and W. Wołyński. Multidimensional economic\nindicators and multivariate functional principal component analysis (MFPCA)\nin a comparative study of countries’competitiveness. Journal of Geographical\nSystems, 24:49–65, 2022.\n[85] J. K. Kueper, M. Speechley, and M. MonteroOdasso.\nThe Alzheimer’s disease\nassessment scale–cognitive subscale (ADASCog):\nModifications and responsiveness\nin predementia\npopulations. A narrative review. Journal of Alzheimer’s Disease,\n63(2):423–444, 2018.\n[86] N. M. Laird and J. H. Ware. Randomeffects\nmodels for longitudinal data. Biometrics,\n38:963–974, 1982.\n[87] K. L. Lanctôt, J. Amatniek, S. AncoliIsrael,\nS. E. Arnold, C. Ballard, J. CohenMansfield,\nZ. Ismail, C. Lyketsos, D. S. Miller, E. Musiek, et al. Neuropsychiatric\nsigns and symptoms of Alzheimer’s disease: New treatment paradigms.\nAlzheimer’s & Dementia: Translational Research & Clinical Interventions,\n3(3):440–449, 2017.\n[88] J. LanteroRodriguez,\nA. Snellman, A. L. Benedet, M. MilàAlomà,\nE. Camporesi,\nL. MontoliuGaya,\nN. J. Ashton, A. Vrillon, T. K. Karikari, J. D. Gispert, et al. Ptau235:\nA novel biomarker for staging preclinical Alzheimer’s disease. EMBO\nmolecular medicine, 13(12):e15098, 2021.\n[89] A. J. Larner. The usage of cognitive screening instruments: Test characteristics and\nsuspected diagnosis. In Cognitive Screening Instruments, pages 219–238. Springer,\nLondon, 2013.\n[90] C. Ledig, A. Schuh, R. Guerrero, R. A. Heckemann, and D. Rueckert. Structural\nbrain imaging in Alzheimer’s disease and mild cognitive impairment: Biomarker\nanalysis and shared morphometry database. Scientific reports, 8(1):1–16, 2018.\n[91] G. Lee, K. Nho, B. Kang, K.A.\nSohn, and D. Kim. Predicting Alzheimer’s disease\nprogression using multimodal\ndeep learning approach. Scientific Reports,\n9(1):1–12, 2019.\n[92] J. C. Lee, S. J. Kim, S. Hong, and Y. Kim. Diagnosis of Alzheimer’s disease\nutilizing amyloid and tau as fluid biomarkers. Experimental & Molecular Medicine,\n51(5):1–10, 2019.\n[93] X. Leng and H.G.\nMüller. Classification using functional data analysis for temporal\ngene expression data. Bioinformatics, 22(1):68–76, 2006.\n[94] A. Li, F. Li, F. Elahifasaee, M. Liu, and L. Zhang. Hippocampal shape and asymmetry\nanalysis by cascaded convolutional neural networks for Alzheimer’s disease\ndiagnosis. Brain Imaging and Behavior, 15(5):2330–2339, 2021.\n[95] B. Li and Q. Yu. Classification of functional data: A segmentation approach. Computational\nStatistics & Data Analysis, 52(10):4790–4800, 2008.\n[96] C. Li, L. Xiao, and S. Luo. Fast covariance estimation for multivariate sparse functional\ndata. Stat, 9(1):e245, 2020.\n[97] D. Li, S. Iddi, W. K. Thompson, M. C. Donohue, and A. D. N. Initiative. Bayesian\nlatent time joint mixed effect models for multicohort longitudinal data. Statistical\nMethods in Medical Research, 28(3):835–845, 2019.\n[98] H. Li, T. Pan, Y. Li, S. Chen, and G. Li. Functional principal component analysis for\nnearinfrared\nspectral data: A case study on Tricholoma matsutakeis. International\nJournal of Food Engineering, 16(8), 2020.\n[99] K. Li and S. Luo. Dynamic prediction of Alzheimer’s disease progression using\nfeatures of multiple longitudinal outcomes and timetoevent\ndata. Statistics in\nMedicine, 38(24):4804–4818, 2019.\n[100] W. Li, X. Lin, and X. Chen. Detecting Alzheimer’s disease based on 4d fMRI: An\nexploration under deep learning framework. Neurocomputing, 388:280–287, 2020.\n[101] X. Li, G. Qi, C. Yu, G. Lian, H. Zheng, S. Wu, T.F.\nYuan, and D. Zhou. Cortical\nplasticity is correlated with cognitive improvement in Alzheimer’s disease\npatients after rTMS treatment. Brain Stimulation, 14(3):503–510, 2021.\n[102] M. P. Lichtenstein, P. Carriba, R. Masgrau, A. Pujol, and E. Galea. Staging antiinflammatory\ntherapy in Alzheimer’s disease. Frontiers in Aging Neuroscience,\n2:142, 2010.\n[103] W. Liggett, L. Cazares, and O. J. Semmes. A look at mass spectral measurement.\nChance, 16(4):24–28, 2003.\n[104] N. Lin, J. Jiang, S. Guo, and M. Xiong. Functional principal component analysis\nand randomized sparse clustering algorithm for medical image analysis. PLoS One,\n10(7):e0132945, 2015.\n[105] M. Liu, D. Cheng, W. Yan, A. D. N. Initiative, et al. Classification of Alzheimer’s\ndisease by combination of convolutional and recurrent neural networks using FDGPET\nimages. Frontiers in Neuroinformatics, 12:35, 2018.\n[106] Y. Liu, L. Tan, H.F.\nWang, Y. Liu, X.K.\nHao, C.C.\nTan, T. Jiang, B. Liu, D.Q.\nZhang, and J.T.\nYu. Multiple effect of APOE genotype on clinical and neuroimaging\nbiomarkers across Alzheimer’s disease spectrum. Molecular Neurobiology,\n53(7):4539–4547, 2016.\n[107] M. Loève. Fonctions aléatoires à décomposition orthogonale exponentielle. La\nRevue Scientifique, 84:159–162, 1946.\n[108] Mayo Clinic Staff. Alzheimer’s stages: How the disease progresses.\nhttps://www.mayoclinic.org/diseases-conditions/alzheimers-disease/\nin-depth/alzheimers-stages/art-20048448. Accessed: 20211101.\n[109] M. Mehdipour Ghazi, M. Nielsen, A. Pai, M. Modat, M. Jorge Cardoso, S. Ourselin,\nand L. Sørensen. Robust parametric modeling of Alzheimer’s disease progression.\nNeuroImage, 225:117460, 2021.\n[110] S. A. Mofrad, A. J. Lundervold, A. Vik, and A. S. Lundervold. Cognitive and MRI\ntrajectories for prediction of Alzheimer’s disease. Scientific Reports, 11(1):1–10,\n2021.\n[111] R. C. Mohs, D. Knopman, R. C. Petersen, S. H. Ferris, C. Ernesto, M. Grundman,\nM. Sano, L. Bieliauskas, D. Geldmacher, C. Clark, et al. Development of cognitive\ninstruments for use in clinical trials of antidementia drugs: Additions to the\nAlzheimer’s disease assessment scale that broaden its scope. Alzheimer Disease\nand Associated Disorders, 1997.\n[112] M. Mojirsheibani and C. Shaw. Classification with incomplete functional covariates.\nStatistics & Probability Letters, 139:40–46, 2018.\n[113] A. Möller, G. Tutz, and J. Gertheiss. Random forests for functional covariates.\nJournal of Chemometrics, 30(12):715–725, 2016.\n[114] H.g.\nMüller. Functional modelling and classification of longitudinal data. Scandinavian\nJournal of Statistics, 32(2):223–240, 2005.\n[115] Z. S. Nasreddine, N. A. Phillips, V. Bédirian, S. Charbonneau, V. Whitehead,\nI. Collin, J. L. Cummings, and H. Chertkow. The Montreal Cognitive Assessment,\nMoCA: A brief screening tool for mild cognitive impairment. Journal of the American\nGeriatrics Society, 53(4):695–699, 2005.\n[116] M. Nguyen, T. He, L. An, D. C. Alexander, J. Feng, B. T. Yeo, A. D. N. Initiative,\net al. Predicting Alzheimer’s disease progression using deep recurrent neural\nnetworks. NeuroImage, 222:117203, 2020.\n[117] NIH National Institute on Aging (NIA). How biomarkers help diagnose dementia.\nhttps://www.nia.nih.gov/health/how-biomarkers-help-diagnose-dementia#\nfuture_biomarkers. Accessed: 20220201.\n[118] NIH National Institute on Aging (NIA). How is alzheimer’s disease treated? https:\n//www.nia.nih.gov/health/how-alzheimers-disease-treated. Accessed: 20220201.\n[119] M. B. T. Noor, N. Z. Zenia, M. S. Kaiser, S. A. Mamun, and M. Mahmud. Application\nof deep learning in detecting neurological disorders from magnetic resonance\nimages: A survey on the detection of Alzheimer’s disease, Parkinson’s disease\nand schizophrenia. Brain Informatics, 7(1):1–21, 2020.\n[120] T. Noori, A. R. Dehpour, A. Sureda, E. SobarzoSanchez,\nand S. Shirooie. Role\nof natural products for the treatment of Alzheimer’s disease. European Journal of\nPharmacology, 898:173974, 2021.\n[121] H.J.\nPark, K. J. Friston, C. Pae, B. Park, and A. Razi. Dynamic effective connectivity\nin resting state fMRI. NeuroImage, 180:594–608, 2018.\n[122] Penn Medicine. The 7 stages of Alzheimer’s disease. https://www.pennmedicine.\norg/updates/blogs/neuroscience-blog/2019/november/stages-of-alzheimers.\nAccessed: 20211101.\n[123] R. C. Petersen. Alzheimer’s disease: Progress in prediction. The Lancet Neurology,\n9(1):4–5, 2010.\n[124] J. Pinheiro and D. Bates. Mixedeffects\nmodels in S and SPLUS.\nSpringer, New\nYork, 2006.\n[125] J. Pinheiro, D. Bates, S. DebRoy, D. Sarkar, and R Core Team. nlme: Linear and\nNonlinear Mixed Effects Models, 2013. R package version 3.1153.\n[126] J. R. Quinlan. Induction of decision trees. Machine Learning, 1(1):81–106, 1986.\n[127] J. R. Quinlan. C4.5: Programs for machine learning. Elsevier, 2014.\n[128] G. D. Rabinovici. Controversy and progress in Alzheimer’s disease —FDA approval\nof Aducanumab. New England Journal of Medicine, 385(9):771–774, 2021.\n[129] J. Ramsay, G. Hooker, and S. Graves. Functional Data Analysis with R and MATLAB.\nSpringer, New York, 2009.\n[130] J. Ramsay and B. W. Silverman. Functional Data Analysis (2 ed.). Springer, New\nYork, 2005.\n[131] M. Ravanelli, P. Brakel, M. Omologo, and Y. Bengio. Light gated recurrent units\nfor speech recognition. IEEE Transactions on Emerging Topics in Computational\nIntelligence, 2(2):92–102, 2018.\n[132] C. Reitz. Alzheimer’s disease and the amyloid cascade hypothesis: A critical review.\nInternational journal of Alzheimer’s disease, 2012:Article ID 369808, 11\npages, 2012.\n[133] K. E. Roach, V. Pedoia, J. J. Lee, T. Popovic, T. M. Link, S. Majumdar, and R. B.\nSouza. Multivariate functional principal component analysis identifies waveform\nfeatures of gait biomechanics related to earlytomoderate\nhip osteoarthritis. Journal\nof Orthopaedic Research®, 39(8):1722–1731, 2021.\n[134] F. Rossi and N. Villa. Support vector machine for functional data classification.\nNeurocomputing, 69(79):\n730–742, 2006.\n[135] I. Saied, T. Arslan, and S. Chandran. Classification of Alzheimer’s disease using\nRF signals and machine learning. IEEE Journal of Electromagnetics, RF and\nMicrowaves in Medicine and Biology, 6(1), 2022.\n[136] A. Sarica, R. Vasta, F. Novellino, M. G. Vaccaro, A. Cerasa, A. Quattrone, A. D. N.\nInitiative, et al. MRI asymmetry index of hippocampal subfields increases through\nthe continuum from the mild cognitive impairment to the Alzheimer’s disease.\nFrontiers in Neuroscience, page 576, 2018.\n[137] S. W. Scheff, D. A. Price, F. A. Schmitt, M. A. Scheff, and E. J. Mufson. Synaptic\nloss in the inferior temporal gyrus in mild cognitive impairment and alzheimer’s\ndisease. Journal of Alzheimer’s Disease, 24(3):547–557, 2011.\n[138] P. Scheltens, D. Leys, F. Barkhof, D. Huglo, H. Weinstein, P. Vermersch, M. Kuiper,\nM. Steinling, E. C. Wolters, and J. Valk. Atrophy of medial temporal lobes on\nMRI in ” probable” Alzheimer’s disease and normal ageing: Diagnostic value and\nneuropsychological correlates. Journal of Neurology, Neurosurgery & Psychiatry,\n55(10):967–972, 1992.\n[139] S. A. Sikkes, E. S. de Langede\nKlerk, Y. A. Pijnenburg, F. Gillissen, R. Romkes,\nD. L. Knol, B. M. Uitdehaag, and P. Scheltens. A new informantbased\nquestionnaire for instrumental activities of daily living in dementia. Alzheimer’s & Dementia,\n8(6):536–543, 2012.\n[140] A. Singleton, M. Farrer, J. Johnson, A. Singleton, S. Hague, J. Kachergus, M. Hulihan,\nT. Peuralinna, A. N. Dutra, S. Lincoln, et al. αsynuclein\nlocus triplication\ncauses Parkinson’s disease. Science, 302(5646):841–842, 2003.\n[141] R. Smith, T. Mukerji, and T. Lupo. Correlating geologic and seismic data with\nunconventional resource production curves using machine learning. Geophysics,\n84(2):O39–O47, 2019.\n[142] T. A. Snijders and R. J. Bosker. Multilevel analysis: An introduction to basic and\nadvanced multilevel modeling (2 ed.). Sage Publications, London, 2011.\n[143] H. Sørensen, J. Goldsmith, and L. M. Sangalli. An introduction with medical applications\nto functional data analysis. Statistics in Medicine, 32(30):5222–5240,\n2013.\n[144] R. A. Sperling, P. S. Aisen, L. A. Beckett, D. A. Bennett, S. Craft, A. M. Fagan,\nT. Iwatsubo, C. R. Jack Jr, J. Kaye, T. J. Montine, et al. Toward defining\nthe preclinical stages of Alzheimer’s disease: Recommendations from the National\nInstitute on AgingAlzheimer’s\nAssociation workgroups on diagnostic guidelines\nfor Alzheimer’s disease. Alzheimer’s & dementia, 7(3):280–292, 2011.\n[145] S. Srivastava, R. Ahmad, and S. K. Khare. Alzheimer’s disease and its treatment\nby different approaches: A review. European Journal of Medicinal Chemistry,\n216:113320, 2021.\n[146] J. E. Storey, J. T. Rowland, D. A. Conforti, and H. G. Dickson. The Rowland universal\ndementia assessment scale (RUDAS): A multicultural cognitive assessment\nscale. International Psychogeriatrics, 16(1):13–31, 2004.\n[147] Y. Su and C.C.\nJ. Kuo. On extended long shortterm\nmemory and dependent bidirectional\nrecurrent neural network. Neurocomputing, 356:151–161, 2019.\n[148] Taiwan Alzheimer Disease Association. 認識失智症. http://www.tada2002.org.\ntw/About/IsntDementia, 04 2021. Accessed: 20210928.\n[149] M. Tanveer, B. Richhariya, R. Khan, A. Rashid, P. Khanna, M. Prasad, and C. Lin.\nMachine learning techniques for the diagnosis of Alzheimer’s disease: A review.\nACM Transactions on Multimedia Computing, Communications, and Applications\n(TOMM), 16(1s):1–35, 2020.\n[150] S. J. Teipel, W. Bayer, G. E. Alexander, Y. Zebuhr, D. Teichberg, L. Kulic, M. B.\nSchapiro, H.J.\nMöller, S. I. Rapoport, and H. Hampel. Progression of Corpus\nCallosum Atrophy in Alzheimer Disease. Archives of Neurology, 59(2):243–248,\n02 2002.\n[151] C. G. Thomas, R. A. Harshman, and R. S. Menon. Noise reduction in BOLDbased\nfMRI using component analysis. Neuroimage, 17(3):1521–1537, 2002.\n[152] M. Torso, M. Bozzali, G. Zamboni, M. Jenkinson, S. A. Chance, and A. D. N.\nInitiative. Detection of Alzheimer’s disease using cortical diffusion tensor imaging.\nHuman Brain Mapping, 42(4):967–977, 2021.\n[153] D. Tosun, Z. Demir, D. P. Veitch, D. Weintraub, P. Aisen, C. R. Jack Jr,\nW. J. Jagust, R. C. Petersen, A. J. Saykin, L. M. Shaw, et al. Contribution of\nAlzheimer’s biomarkers and risk factors to cognitive impairment and decline across\nthe Alzheimer’s disease continuum. Alzheimer’s & Dementia, 2021.\n[154] G. Van Rossum and F. L. Drake Jr. Python reference manual. Centrum voor\nWiskunde en Informatica Amsterdam, 1995.\n[155] M. Vernooij, F. Pizzini, R. Schmidt, M. Smits, T. Yousry, N. Bargallo, G. Frisoni,\nS. Haller, and F. Barkhof. Dementia imaging in clinical practice: A europeanwide\nsurvey of 193 centres and conclusions by the ESNR working group. Neuroradiology,\n61(6):633–642, 2019.\n[156] R. Viviani, G. Grön, and M. Spitzer. Functional principal component analysis of\nfMRI data. Human brain mapping, 24(2):109–129, 2005.\n[157] M. Walterfang, E. Luders, J. C. Looi, P. Rajagopalan, D. Velakoulis, P. M. Thompson,\nO. Lindberg, P. Östberg, L. E. Nordin, L. Svensson, et al. Shape analysis of\nthe corpus callosum in Alzheimer’s disease and frontotemporal lobar degeneration\nsubtypes. Journal of Alzheimer’s Disease, 40(4):897–906, 2014.\n[158] J.L.\nWang, J.M.\nChiou, and H.G.\nMüller. Functional data analysis. Annual Review\nof Statistics and Its Application, 3:257–295, 2016.\n[159] L. Wang, Y. Zang, Y. He, M. Liang, X. Zhang, L. Tian, T. Wu, T. Jiang, and K. Li.\nChanges in hippocampal connectivity in the early stages of Alzheimer’s disease:\nEvidence from resting state fMRI. Neuroimage, 31(2):496–504, 2006.\n[160] Y. Wei, G. Xiao, H. Deng, H. Chen, M. Tong, G. Zhao, and Q. Liu. Hyperspectral\nimage classification using FPCAbased\nkernel extreme learning machine. Optik,\n126(23):3942–3948, 2015.\n[161] R. K. Wong, Y. Li, and Z. Zhu. Partially linear functional additive models for\nmultivariate functional data. Journal of the American Statistical Association,\n114(525):406–418, 2019.\n[162] World Health Organization. Dementia. https://www.who.int/news-room/\nfact-sheets/detail/dementia. Accessed: 2021-09-28.\n[163] World Health Organization. Dementia: a public health priority. https://www.who.\nint/publications/i/item/dementia-a-public-health-priority. Accessed: 2021-09-28.\n[164] World Health Organization. The top 10 causes of death. https://www.who.int/\nnews-room/fact-sheets/detail/the-top-10-causes-of-death. Accessed: 2021-09-28.\n[165] Y. Wu and Y. Liu. Functional robust support vector machines for sparse and\nirregular longitudinal data. Journal of computational and Graphical Statistics,\n22(2):379–395, 2013.\n[166] S. Xie. Wavelet power spectral domain functional principal component analysis for\nfeature extraction of epileptic EEGs. Computation, 9(7):78, 2021.\n[167] F. Xue, F. Tan, Z. Ye, J. Chen, and Y. Wei. Spectralspatial\nclassification of hyperspectral\nimage using improved functional principal component analysis. IEEE\nGeoscience and Remote Sensing Letters, 19:1–5, 2021.\n[168] B. Yang, H. Yu, M. Xing, R. He, R. Liang, and L. Zhou. The relationship between\ncognition and depressive symptoms, and factors modifying this association,\nin Alzheimer’s disease: A multivariate multilevel model. Archives of Gerontology\nand Geriatrics, 72:25–31, 2017.\n[169] L. Yang, J. Yan, X. Jin, Y. Jin, W. Yu, S. Xu, and H. Wu. Screening for dementia\nin older adults: Comparison of MiniMental\nState Examination, MiniCog,\nClock\nDrawing Test and AD8. PLOS ONE, 11(12):1–9, 12 2016.\n[170] F. Yao, E. Lei, and Y. Wu. Effective dimension reduction for sparse functional data.\nBiometrika, 102(2):421–437, 2015.\n[171] F. Yao, H.G.\nMüller, and J.L.\nWang. Functional data analysis for sparse longitudinal\ndata. Journal of the American statistical association, 100(470):577–590,\n2005.\n[172] F. Yao, Y. Wu, and J. Zou. Probabilityenhanced\neffective dimension reduction for\nclassifying sparse functional data. Test, 25(1):1–22, 2016.\n[173] L. Zhang, M. Wang, M. Liu, and D. Zhang. A survey on deep learning for\nneuroimagingbased\nbrain disorder analysis. Frontiers in Neuroscience, page 779,\n2020.\n[174] 台灣神經學學會Taiwan Neurological Society. 台灣神經學學會會訊2020 年\n01 月第80 期. http://www.neuro.org.tw/files/newsletter/080.pdf. Accessed:\n2021-09-28.\n[175] 衛生福利部Ministry of Health and Welfare. 失智症防治照護政策綱\n領暨行動方案2.0(含工作項目)(2021 年版). https://1966.gov.tw/LTC/\ncp-4020-42469-201.html. Accessed: 2021-09-28.\n[176] 衛生福利部中央健康保險署National Health Insurance Administration, Ministry\nof Health and Welfare. 最新版藥品給付規定內容\n第1 節神經系統藥\n物drugs acting on the nervous system. https://www.nhi.gov.tw/Content_List.\naspx?n=E70D4F1BD029DC37&topn=5FE8C9FEAE863B46. Update: 20220224,\nAccessed: 2022-03-02.\n[177] 衛生福利部統計處Department of Statistics, Ministry of Health and Welfare.\n國際失智症日衛生福利統計通報. https://www.mohw.gov.tw/\ndl-71799-1d824fee-a486-4504-9c7d-5d819c6848b2.html. Accessed: 2021-09-28. zh_TW dc.identifier.doi (DOI) 10.6814/NCCU202201025 en_US
