失智症（Dementia）根據嚴重程度可分為阿茲海默症（Alzheimer’s Disease, AD）和輕度認知障礙（Mild Cognitive Impairment, MCI）。雖然藥物治療可以緩解失智症的症狀，但目前尚無法根治，且若延遲診斷，治療效果可能會下降。因此，早期診斷和治療對於失智症至患者關重要。在過去利用機器學習與腦電圖（Electroencephalogram, EEG）結合應用於失智症分類時，通常會利用傳統頻帶計算特徵，但由於傳統頻帶在失智症分類中缺乏泛化性，因此透過傳統頻帶計算的特徵訓練分類時，此分類器在測試集數據進行分類時容易有準確率下降的表現。本研究結合了自我迴歸模型數據驅動頻帶邊界方法與腦電圖，以識別 AD、MCI 和健康對照組（Health Controls, HC）的泛化性頻帶，並訓練二元分類器，研究目的是透過計算泛化性頻帶的能量作為訓練分類器的特徵，並將其結果與固定的傳統頻帶的結果進行比較。在分類器方面，使用了最小絕對值收斂與選擇算子邏輯迴歸（Least Absolute Shrinkage and Selection Operator, Lasso）以及分組最小絕對值收斂與選擇算子邏輯迴歸(Group Least Absolute Shrinkage and Selection Operator, Group Lasso），其中 Group Lasso 能夠進行組別特徵選擇，從而幫助識別重要的腦區。結果顯示在 AD-HC 分類中，自我迴歸模型數據驅動的頻帶邊界方法所找到的頻帶準確率為 0.75，高於傳統頻帶訓練分類器準確率為0.42的表現。在 HC-MCI 分類中，自我迴歸模型數據驅動方法找到的頻帶的準確率僅略高於傳統頻帶訓練分類器，其準確率約為 0.75。此外，Group Lasso的腦區選擇結果顯示，額葉(Frontal)和右顳(Right Temporal)在 HC-MCI 分類中被認為是重要腦區，並且在找到的頻帶與傳統頻帶結果中展現一致性。雖然本研究在 AD-MCI 的分類結果未達預期，但在 AD-HC 和 HC-MCI 的分類中取得了不錯的準確率。這表明通過找到泛化性頻帶進行失智症分類，雖然尚未達到卓越的表現，但在 AD-HC 分類中顯示出明顯的差異，這為未來的方法改進和潛力提供了提升的空間。

Although medication can alleviate symptoms of dementia, there is currently no cure, and delayed diagnosis may reduce treatment effectiveness. Therefore, early diagnosis and treatment are crucial for dementia patients. This study integrates a data-driven frequency band boundary method based on Autoregressive Models (AR model) with EEG to identify generalized frequency bands for classifying Alzheimer’s Disease (AD), Mild Cognitive Impairment (MCI), and Healthy Controls (HC). The objective is to use the power of these generalized frequency bands as features for training classifiers, and to compare the results with those obtained from fixed conventional frequency bands (CFB). For the classifiers, this study employed Least Absolute Shrinkage and Selection Operator (Lasso) logistic regression and Group Least Absolute Shrinkage and Selection Operator (Group Lasso) logistic regression; the latter enables group feature selection to identify important brain regions. The classification accuracy for AD-HC classification is 0.75, which is significantly better than that of models trained with fixed conventional frequency band features. In the HC-MCI classification, an accuracy of 0.75 can still be achieved using only data from the frontal and right temporal regions.

[1]Prince, M.; Wimo, A.; Guerchet, M.; Ali, G.-C.; Wu, Y.-T.; Prina, M. “World Alzheimer Report 2015, the Global Impact of Dementia: An Analysis of Prevalence, Incidence, Cost and Trends” Alzheimer’s Disease International: London, UK, 2015; p. 87
[2]B. Oltu, M. F. Akşahin, and S. Kibaroğlu, “A novel electroencephalography based approach for Alzheimer’s disease and mild cognitive impairment detection,” Biomedical Signal Processing and Control, vol. 63, p. 102223, Jan. 2021.
[3]S. Sharma and P. K. Mandal, “A Comprehensive Report on Machine Learning-based Early Detection of Alzheimer’s Disease using Multi-modal Neuroimaging Data,” ACM Comput. Surv., vol. 55, no. 2, pp. 1–44, Feb. 2023.
[4]E. Niedermeyer, “Alpha rhythms as physiological and abnormal phenomena,” International Journal of Psychophysiology, vol. 26, no. 1–3, pp. 31–49, Jun. 1997.
[5]M. Crous-Bou, C. Minguillón, N. Gramunt, and J. L. Molinuevo, “Alzheimer’s disease prevention: from risk factors to early intervention,” Alz Res Therapy, vol. 9, no. 1, p. 71, Dec. 2017.
[6]T.-T. Trinh, C.-F. Tsai, Y.-T. Hsiao, C.-Y. Lee, C.-T. Wu, and Y.-H. Liu, “Identifying Individuals With Mild Cognitive Impairment Using Working Memory-Induced Intra-Subject Variability of Resting-State EEGs,” Front. Comput. Neurosci., vol. 15, p. 700467, Aug. 2021.
[7]Xiao, J., J. Li, J. Wang, et al. 2023. “2023 China Alzheimer's Disease: Facts and Figures.” Human Brain 2, no. 3: 1–13.
[8]A. Nordlund, “The Goteborg MCI study: mild cognitive impairment is a heterogeneous condition,” Journal of Neurology, Neurosurgery & Psychiatry, vol. 76, no. 11, pp. 1485–1490, Nov. 2005.
[9]P. Mohankumar, J. Ajayan, T. Mohanraj, and R. Yasodharan, “Recent developments in biosensors for healthcare and biomedical applications: A review,” Measurement, vol. 167, p. 108293, Jan. 2021.
[10]G. Kuslansky et al., “Detecting dementia with the Hopkins Verbal Learning Test and the Mini-Mental State Examination,” Archives of Clinical Neuropsychology, 2004.
[11]M. Ouchani, S. Gharibzadeh, M. Jamshidi, and M. Amini, “A Review of Methods of Diagnosis and Complexity Analysis of Alzheimer’s Disease Using EEG Signals,” BioMed Research International, vol. 2021, no. 1, p. 5425569, 2021.
[12]D. P. Subha, P. K. Joseph, R. Acharya U, and C. M. Lim, “EEG Signal Analysis: A Survey,” J Med Syst, vol. 34, no. 2, pp. 195–212, Apr. 2010.
[13]J. Li, X. Li, F. Chen, W. Li, J. Chen, and B. Zhang, “Studying the Alzheimer’s disease continuum using EEG and fMRI in single-modality and multi-modality settings,” Reviews in the Neurosciences, vol. 35, no. 4, pp. 373–386, Jun. 2024.
[14]J. P. Amezquita-Sanchez, N. Mammone, F. C. Morabito, and H. Adeli, “A New dispersion entropy and fuzzy logic system methodology for automated classification of dementia stages using electroencephalograms,” Clinical Neurology and Neurosurgery, vol. 201, p. 106446, Feb. 2021.
[15]J. Jeong, “EEG dynamics in patients with Alzheimer’s disease,” Clinical Neurophysiology, vol. 115, no. 7, pp. 1490–1505, Jul. 2004.
[16]J. Dauwels, F. Vialatte, and A. Cichocki, “Diagnosis of Alzheimer’s Disease from EEG Signals: Where Are We Standing?” Current Alzheimer Res., vol. 7, pp. 487–505, 2010.
[17]J. Jiang et al., “A Novel Detection Tool for Mild Cognitive Impairment Patients Based on Eye Movement and Electroencephalogram,” JAD, vol. 72, no. 2, pp. 389–399, Nov. 2019.
[18]M. K. Siddiqui, R. Morales-Menendez, X. Huang, and N. Hussain, “A review of epileptic seizure detection using machine learning classifiers,” Brain Inf., vol. 7, no. 1, p. 5, Dec. 2020.
[19]L. Hu and Z. Zhang, Eds., EEG Signal Processing and Feature Extraction. Singapore: Springer Singapore, 2019.
[20]A. Modir, S. Shamekhi, and P. Ghaderyan, “A systematic review and methodological analysis of EEG-based biomarkers of Alzheimer’s disease,” Measurement, vol. 220, p. 113274, Oct. 2023.
[21]S. Gaubert, M. Houot, F. Raimondo, M. Ansart, M.-C. Corsi, L. Naccache, J. D. Sitt, M.-O. Habert, B. Dubois, F. D. V. Fallani et al., “A machine learning approach to screen for preclinical alzheimer’s disease,” Neurobiology of Aging, vol. 105, pp. 205–216, 2021.
[22]L. Tait et al., “EEG microstate complexity for aiding early diagnosis of Alzheimer’s disease,” Sci Rep, vol. 10, no. 1, p. 17627, Oct. 2020.
[23]Y. Miao et al., “Dynamic theta/beta ratio of clinical EEG in Alzheimer’s disease,” Journal of Neuroscience Methods, vol. 359, p. 109219, Jul. 2021.
[24]S. Siuly et al., “A New Framework for Automatic Detection of Patients With Mild Cognitive Impairment Using Resting-State EEG Signals,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 28, no. 9, pp. 1966–1976, Sep. 2020.
[25]Y. Jiang et al., “Memory-Related Frontal Brainwaves Predict Transition to Mild Cognitive Impairment in Healthy Older Individuals Five Years Before Diagnosis,” Journal of Alzheimer’s Disease, vol. 79, no. 2, pp. 531–541, Jan. 2021.
[26]V. Kulkarni, Y. Joshi, and R. Manthalkar, “Sensory-Motor Cortex Signal Classification for Rehabilitation Using EEG Signal,” 2021, pp. 273–282.
[27]Y.-T. Hsiao, C.-T. Wu, C.-F. Tsai, Y.-H. Liu, T.-T. Trinh, and C.-Y. Lee, “EEG-Based Classification Between Individuals with Mild Cognitive Impairment and Healthy Controls Using Conformal Kernel-Based Fuzzy Support Vector Machine,” Int. J. Fuzzy Syst., vol. 23, no. 8, pp. 2432–2448, Nov. 2021.
[28]K. D. Tzimourta et al., “EEG Window Length Evaluation for the Detection of Alzheimer’s Disease over Different Brain Regions,” Brain Sciences, vol. 9, no. 4, p. 81, Apr. 2019.
[29]X. Chen, X. Qi, Z. Wang, C. Cui, B. Wu, and Y. Yang, “Fault diagnosis of rolling bearing using marine predators algorithm-based support vector machine and topology learning and out-of-sample embedding,” Measurement, vol. 176, p. 109116, May 2021,.
[30]M. Aviles, L. M. Sánchez-Reyes, J. M. Álvarez-Alvarado, and J. Rodríguez-Reséndiz, “Machine and Deep Learning Trends in EEG-Based Detection and Diagnosis of Alzheimer’s Disease: A Systematic Review,” Eng, vol. 5, no. 3, pp. 1464–1484, Jul. 2024.
[31]S. K. Parmar, O. A. Ramwala, and C. N. Paunwala, “Performance Evaluation of SVM with Non-Linear Kernels for EEG-based Dyslexia Detection,” in 2021 IEEE 9th Region 10 Humanitarian Technology Conference (R10-HTC), Sep. 2021, pp. 1–6.
[32]N. N. Kulkarni and V. K. Bairagi, “Extracting Salient Features for EEG-based Diagnosis of Alzheimer’s Disease Using Support Vector Machine Classifier,” IETE Journal of Research, vol. 63, no. 1, pp. 11–22, Jan. 2017.
[33]P. Durongbhan et al., “A Dementia Classification Framework Using Frequency and Time-Frequency Features Based on EEG Signals,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 27, no. 5, pp. 826–835, May 2019.
[34]S. Khatun, B. I. Morshed, and G. M. Bidelman, “A Single-Channel EEG-Based Approach to Detect Mild Cognitive Impairment via Speech-Evoked Brain Responses,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 27, no. 5, pp. 1063–1070, May 2019.
[35]C. Hima et al., ‘‘A review on brainwave therapy,’’ World J. Pharmaceutical Sci., pp. 59–66, 2020.
[36]P. A. M. Kanda, E. F. Oliveira, and F. J. Fraga, “EEG epochs with less alpha rhythm improve discrimination of mild Alzheimer’s,” Computer Methods and Programs in Biomedicine, vol. 138, pp. 13–22, Jan. 2017.
[37]C. Babiloni et al., “Abnormal fronto‐parietal coupling of brain rhythms in mild Alzheimer’s disease: a multicentric EEG study,” Eur J of Neuroscience, vol. 19, no. 9, pp. 2583–2590, May 2004.
[38]Y. A. L. Pijnenburg, Y. Vd Made, A. M. Van Cappellen Van Walsum, D. L. Knol, Ph. Scheltens, and C. J. Stam, “EEG synchronization likelihood in mild cognitive impairment and Alzheimer’s disease during a working memory task,” Clinical Neurophysiology, vol. 115, no. 6, pp. 1332–1339, Jun. 2004.
[39]郭邵頎,“共同空間型樣法與沿伸之數據驅動頻帶應用於癲癇預測.” 碩士論文, 國立成功大學機械工程學系, 臺灣臺南市, 2024.
[40]S. Talebi, J. Waczak, B. A. Fernando, A. Sridhar, and D. J. Lary, “Data-Driven EEG Band Discovery with Decision Trees,” Sensors, vol. 22, no. 8, p. 3048, Apr. 2022.
[41]M. X. Cohen, “A data-driven method to identify frequency boundaries in multichannel electrophysiology data,” Journal of Neuroscience Methods, vol. 347, p. 108949, Jan. 2021.
[42]莊昱呈,“拉蓋多爾-沃爾泰拉自我回歸模型用於阿茲海默症、輕度認知障礙和健康對照組的分類.” 碩士論文, 國立成功大學機械工程學系, 臺灣臺南市, 2023.
[43]L.-F. Cai, C.-L. Wang, H.-W. Chen, H. Qian, Z.-Y. Lin, and X.-C. Zhang, “Hydrogen bonding-based self-assembly technology for high-performance melt blending TPU/PA6 polymers,” Appl Nanosci, vol. 10, no. 1, pp. 51–59, Jan. 2020.
[44]A. Isaksson, A. Wennberg, and L. H. Zetterberg, “Computer analysis of EEG signals with parametric models,” Proc. IEEE, vol. 69, no. 4, pp. 451–461, 1981.
[45]B. Y. R. Shebata, “Selection of the order of an autoregressive model by Akaike’s information criterion” Biometrika, vol. 63, no. 1, pp. 117–126, 1976.
[46]M. Ester, H.-P. Kriegel, and X. Xu, “A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise” in Proc. Kdd, vol. 96, 1996, pp. 226–231.
[47]L. Meier, S. Van De Geer, and P. Bühlmann, “The group lasso for logistic regression,” Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 70, no. 1, pp. 53–71, 2008.
[48]Y. Yang and H. Zou, “A fast unified algorithm for solving group-lasso penalize learning problems,” Stat Comput, vol. 25, no. 6, pp. 1129–1141, Nov. 2015.
[49]M. Sivakumar, S. Parthasarathy, and T. Padmapriya, “Trade-off between training and testing ratio in machine learning for medical image processing,” PeerJ Computer Science, vol. 10, p. e2245, Sep. 2024.
[50]P. M. Rossini et al., “Is it possible to automatically distinguish resting EEG data of normal elderly vs. mild cognitive impairment subjects with high degree of accuracy?,” Clinical Neurophysiology, vol. 119, no. 7, pp. 1534–1545, Jul. 2008.
[51]Y.-T. Hsiao, C.-F. Tsai, C.-T. Wu, T.-T. Trinh, C.-Y. Lee, and Y.-H. Liu, “MCI Detection Using Kernel Eigen-Relative-Power Features of EEG Signals,” Actuators, vol. 10, no. 7, p. 152, Jul. 2021.