臺灣博碩士論文加值系統

失智症人口逐年攀升，影響到許多家庭及醫療系統。患者須及早治療才能延緩其惡化的速度，而傳統的診斷方法需要依賴專業醫護人員的經驗且費時，無法應付龐大的需求。因此我們提出一個以語句中的語音特徵做為輸入的失智辨識模型，來解決這樣的問題。我們將Mandarin_Lu及Aishell語音庫組成訓練資料，並嘗試解決資料不平衡的問題，以長短期記憶神經網路(Long-short term memory)作為基礎，訓練出一個失智辨識模型，在兩種語音庫組成的測試集上達到98%的準確率。

The population of people who suffer from dementia is increasing year by year. Early detection of dementia is very important for the patient. However, traditional testing tools rely on experienced doctor to conduct and take long time. Therefore, this thesis presents an approach to detect dementia automatically through acoustic features. The proposed method employs LSTM recurrent neural network on MFCC features extracted from spoken utterances to build a predictive model. We train the model on utterances which come from two speech corpora (Mandarin_Lu and Aishell) and deal with imbalanced data. Our model achieves an accuracy of 98% on test set.

口試委員審定書 i
致謝 ii
中文摘要 iii
ABSTRACT iv
目錄 v
圖目錄 vii
表目錄 ix
第 1 章 緒論 1
1.1 研究動機及目的 1
1.2 文獻回顧 1
1.3 本論文貢獻 3
1.4 章節概要 3
第 2 章 背景知識 5
2.1 失智症 5
2.2 臨床失智症量測評估工具 5
2.2.1 MMSE 5
2.2.2 CDR 6
2.3 類神經網路 7
2.3.1 遞歸類神經網路 10
2.3.2 長短期記憶神經網路 11
第 3 章 系統架構與特徵擷取 13
3.1 系統架構 13
3.2 特徵擷取 14
3.3 特徵標準化 15
第 4 章 資料來源及處理 16
4.1 資料來源 16
4.2 資料分析與處理 19
4.2.1 語句長度分析 19
4.2.2 噪音處理 20
第 5 章 實驗與結果討論 23
5.1 實驗流程與環境設定 23
5.2 加入噪音之影響 25
5.3 模型架構比較 27
5.3.1 LSTM輸出層 27
5.3.2 LSTM層數 31
5.4 模型對不同資料集之強健性 33
第 6 章 結論 35
參考文獻 36
附錄 39

[1]S.Sarraf, D. D.Desouza, and J.Anderson, “DeepAD : Alzheimer ’ s Disease Classification via Deep Convolutional Neural Networks using MRI and fMRI,” pp. 1–32, 2016.
[2]A.Payan and G.Montana, “Predicting Alzheimer ’ s disease : a neuroimaging study with 3D convolutional neural networks,” pp. 1–9, 2015.
[3]M.Lu, “Attention-based Deep Multiple Instance Learning for MCI discriminative patch localization and diagnosis on brain MRI,” 國立台灣大學碩士論文, July, 2019.
[4]S.Olubolu, O.Id, J. S.Wong, and C. P.Wong, “Deep language space neural network for classifying mild cognitive impairment and Alzheimer-type dementia,” pp. 1–15, 2018.
[5]T.Niu, M.Bansal, and S.Karlekar, “Detecting Linguistic Characteristics of Alzheimer’s Dementia by Interpreting Neural Models,” 2014.
[6]Y.Liu, “A Classifier for Alzheimer ’ s Disease Evaluation Based on Monologue Transcription Data,” 國立台灣大學碩士論文, July, 2018.
[7]W.Jarrold, M. L.Gorno-tempini, and J.Ogar, “Aided Diagnosis of Dementia Type through Computer-Based Analysis of Spontaneous Speech,” pp. 27–37, 2014.
[8]A.König et al., “Automatic speech analysis for the assessment of patients with predementia and Alzheimer’s disease,” Alzheimer’s Dement. Diagnosis, Assess. Dis. Monit., vol. 1, no. 1, pp. 112–124, 2015.
[9]S.Kato, A.Homma, T.Sakuma, and M.Nakamura, “Detection of Mild Alzheimer ’ s Disease and Mild Cognitive Impairment from Elderly Speech : binary discrimination using logistic regression,” pp. 5569–5572, 2015.
[10]M. F.Folstein, S. E.Folstein, and P. R.McHugh, “‘Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician,” J. Psychiatr. Res., 1975.
[11]C. P.Hughes, L.Berg, W. L.Danziger, L. A.Coben, and R. L.Martin, “A new clinical scale for the staging of dementia,” Br. J. Psychiatry, 1982.
[12]S.Borson, J.Scanlan, M.Brush, P.Vitaliano, and A.Dokmak, “The mini-cog: A cognitive ‘vital signs’ measure for dementia screening in multi-lingual elderly,” Int. J. Geriatr. Psychiatry, 2000.
[13]E.Pfeiffer, “A Short Portable Mental Status Questionnaire for the Assessment of Organic Brain Deficit in Elderly Patients,” J. Am. Geriatr. Soc., 1975.
[14]J. E.Galvin et al., “The AD8: A brief informant interview to detect dementia,” Neurology, 2005.
[15]D. E.Rumelhart, G. E.Hinton, and R. J.Williams, “Learning representations by back-propagating errors,” Nature, 1986.
[16]S.Hochreiter and J.Schmidhuber, “Long Short-Term Memory,” Neural Comput., 1997.
[17]B.MacWhinney, D.Fromm, M.Forbes, and A.Holland, “Aphasiabank: Methods for studying discourse,” Aphasiology, 2011.
[18]H.Bu, J.Du, X.Na, B.Wu, and H.Zheng, “AISHELL-1: An open-source Mandarin speech corpus and a speech recognition baseline,” in 2017 20th Conference of the Oriental Chapter of International Committee for Coordination and Standardization of Speech Databases and Assessment Techniques, O-COCOSDA 2017, 2018.
[19]E. F.Kaplan, H.Goodglass, and S.Weintraub, “The Boston Naming Test.,” Philadelphia Lea Febiger., 1983.
[20]F.Eyben, M.Wollmer, and B.Schuller, “openSMILE - The Munich Versatile and Fast Open-Source Audio Feature Extractor Categories and Subject Descriptors,” in ACM Multimedia, 2010.
[21]F.Chollet, “Keras: Deep Learning library for Theano and TensorFlow,” GitHub Repos., 2015.