臺灣博碩士論文加值系統

隨著台灣步入高齡化社會後，老年人口比例急遽增加，中風患者也因此而增加，在中風後要復健的需求已成為常態。但長期的復健行為容易讓患者產生疲勞感，進而影響復健成效，甚至可能會產生負面效果導致復健失效。因此如何即時診斷心理疲勞是迫切需要被深入研究的議題。在過往針對心理疲勞使用腦電訊號的研究已行之有年，在時域以及頻域上的特徵皆有所斬獲，但僅是對狀態的辨別，而在腦區間的關聯較少被提及討論。本研究提出一種可適用於不同個體的通道選擇方法來辨識心理疲勞及探討腦區間連結程度的轉移。
本研究以TloadDback範式建立心理疲勞誘發實驗來獲取受試者之EEG圖(Electroencephalography，EEG)數據集。為解決雜訊問題，本研究改良強化奇異值分解(Enhanced Singular Value Decomposition, ESVD)的數據重構方法來將EEG訊號重構以獲得更高訊號乾淨度，並經由信噪比(提高9.32dB)及絕對平均平方誤差(0.0478)來驗證有效性。為了解決辨識心理疲勞中存在個體差異的問題，本研究提出一種新自適應EEG通道相位同步法，其基於ESVD將加權相位遲滯指數(weighted Phase Lag Index, wPLI)進行次數及分數的綜合最佳排名，因此可針對不同個體更專注於挑選具有重要特徵的通道。結果顯示，利用此方法可全面提高傳統機器學習方法的分類準確率。為了瞭解個體在發生疲勞與非疲勞時的腦區狀態變化，本研究將基於ESVD的wPLI來觀測腦區相關性的轉換狀態。結果顯示，腦區通道間高相位同步性能有效區分疲勞和非疲勞狀態。此外，無論有無發生疲勞，大部分受試者的左右腦有高度不對稱性，並且會基於左前半腦與其他腦區進行連結，同時也發現右後半腦的活躍程度相對低。

As Taiwan entered an aging society, the proportion of the elderly population had increased rapidly, and the number of stroke patients had also increased. The need for rehabilitation after a stroke had become the norm. However, long-term rehabilitation behavior could easily cause patients to feel fatigued, which affected the effectiveness of rehabilitation, and may have even had negative effects and led to rehabilitation failure. Therefore, how to diagnose mental fatigue immediately was an issue that needed to be studied in depth. EEG features of mental fatigue in the time domain and frequency domain had been discussed, but only the identification of states. The correlation between brain regions were less discussed. This study proposed a channel selection method that could be applied to different individuals to identify mental fatigue and explore the transfer of brain-to-brain connectivity. This study used the TloadDback paradigm to establish a mental fatigue induction experiment to obtain the subjects' Electroencephalography (EEG) data set. To solve the problem of noise and artifacts, the researcher introduced and improved a data reconstruction method based on Enhanced Singular Value Decomposition (ESVD) to reconstruct the EEG signal and obtain higher signal cleanliness. The results showed that ESVD improved the signal-to-noise ratio by 9.32dB and absolute mean squared error (0.0478) to verify the effectiveness. To solve the problem of individual differences in recognition of mental fatigue, this study proposed a new adaptive EEG channel phase synchronization method, which was based on the comprehensive optimization ranking of the ESVD-based weighted phase lag index (wPLI) times and scores. So that channels with important characteristics could be more focused on picking different individuals. The results showed that using this method could comprehensively improve the classification accuracy of traditional machine learning methods. To understand the changes in the brain region state of an individual when fatigue and non-fatigue occurred, this study observed the transition state of brain region correlation using ESVD-based wPLI. The results showed high phase synchronization performance between brain area channels could effectively distinguish fatigue and non-fatigue states. In addition, regardless of whether fatigue occurred, most subjects had a high asymmetry between the left and right brains, and connections with other brain areas were based on the left front hemisphere. This study also found that the right back hemisphere's activities were relatively low.

摘要 i
Abstract ii
目錄 iv
圖目錄 vi
表目錄 ix
第一章 緒論 1
1.1高齡化社會及中風患者引發問題 1
1.2心理疲勞的影響與重要性 2
第二章 文獻回顧 5
2.1心理疲勞誘發實驗方法 5
2.2心理疲勞辨識技術 8
2.3文獻回顧總結 12
2.4研究目的、貢獻與原創性 12
第三章 研究方法 14
3.1心理疲勞誘發實驗 14
3.1.1 TloadDback範式 14
3.1.2實驗設計 15
3.2強化奇異值分解 19
3.3自適應EEG通道相位同步法 23
第四章 結果與討論 25
4.1主觀疲勞量表結果 25
4.2數據評估結果 26
4.2.1降噪方法 26
4.2.2頻帶功率 31
4.2.3分類器 40
4.2.4自適應EEG通道相位同步分析 43
4.3討論 55
第五章 結論與未來展望 57
5.1結論 57
5.2未來展望 57
References 59
附錄 66
附件 1.IRB同意書 66
附件 2.FAS問卷結果 67
附件 3.wPLI 60分鐘前10排名 72

[1]Tong, R. K., Ng, M. F., & Li, L. S. “Effectiveness of gait training using an electromechanical gait trainer, with and without functional electric stimulation, in subacute stroke： a randomized controlled trial”. Arch Phys Med Rehabil, 87(10)(2006), 1298-1304. https://doi.org/10.1016/j.apmr.2006.06.016.
[2]三階段人口(占總人口比率)，國家發展委員會，https://www.ndc.gov.tw/Content_List.aspx?n=0F11EF2482E76C5，檢自2023.06.30
[3]Rowat, A., Pollock, A., St George, B., Cowey, E., Booth, J., & Lawrence, M.“Top 10 research priorities relating to stroke nursing： A rigorous approach to establish a national nurse‐led research agenda”, Journal of Advanced Nursing, 72(11) (2016), 2831-2843. https://doi.org/10.1111/jan.13048.
[4]Cumming, T.B., Packer, M., Kramer, S.F., English, C. “The prevalence of fatigue after stroke： a systematic review and meta-analysis”, Int. J. Stroke 11 (9) (2016), 968 – 977. https://doi.org/10.1177/1747493016669861.
[5]Vuletić V, Lezaić Z, Morović S. Post-stroke fatigue. Acta Clin Croat. 506(3) (2011), 341-344. https://hrcak.srce.hr/84092.
[6]de Groot, M.H., Phillips, S.J., Eskes, G.A. Fatigue associated with stroke and other neurologic conditions： implications for stroke rehabilitation. Arch. Phys. Med. Rehabil. 84 (11) (2003), 1714 – 1720. https://doi.org/10.1053/S0003-9993(03)00346-0.
[7]de Bruijn, M.A.A.M., Synhaeve, N.E., van Rijsbergen, M.W.A., de Leeuw, F.-E., Mark, R. E., Jansen, B.P.W., de Kort, P.L.M. Quality of life after young ischemic stroke of mild severity is mainly influenced by psychological factors. J. Stroke Cerebrovasc. Dis. 24 (10) (2015), 2183 – 2188. https://doi.org/10.1016/j.jstrokecerebrovasdis.2015.04.040.
[8]Glader, E.-L., Stegmayr, B., Asplund, K. Poststroke fatigue： a 2-year follow-up study of stroke patients in Sweden. Stroke 33 (5) (2002), 1327 – 1333. https://doi.org/10.1161/01.STR.0000014248.28711.D6.
[9]Kristine M. Ulrichsen et al, Structural brain disconnectivity mapping of post-stroke fatigue, NeuroImag：Clinical, 30(2021). https://doi.org/10.1016/j.nicl.2021.102635.
[10]Darnai, G., Matuz, A., Alhour, H. A., Perlaki, G., Orsi, G., Arató, Á., Szente, A., Áfra, E., Nagy, S., Janszky, J. & Csathó, Á. The neural correlates of mental fatigue and reward processing： A task-based fMRI study. NeuroImage, 265(2023), 119812. https://doi.org/10.1016/j.neuroimage.2022.119812.
[11]Mehmood, I., Li, H., Umer, W., Arsalan, A., Shakeel, M. S., & Anwer, S. Validity of facial features’ geometric measurements for real-time assessment of mental fatigue in construction equipment operators. Advanced Engineering Informatics, 54 (2022), 101777. https://doi.org/10.1016/j.aei.2022.101777.
[12]Ponce-Bordón, J. C., García-Calvo, T., López-Gajardo, M. A., Díaz, J., & González-Ponce, I. How does the manipulation of time pressure during soccer tasks influence physical load and mental fatigue? Psychology of Sport and Exercise, (2022), 102253. https://doi.org/10.1016/j.psychsport.2022.102253.
[13]Ye, C., Yin, Z., Zhao, M., Tian, Y., & Sun, Z. Identification of mental fatigue levels in a language understanding task based on multi-domain EEG features and an ensemble convolutional neural network. Biomedical Signal Processing and Control, 72 (2022), 103360. https://doi.org/10.1016/j.bspc.2021.103360.
[14]Melo, H. M., Nascimento, L. M., de Bem Alves, A. C., Walz, R., & Takase, E. N2 event-related potential component is associated with cardiac autonomic tone regulation during mental fatigue. Physiology & Behavior, 241 (2021), 113591. https://doi.org/10.1016/j.physbeh.2021.113591.
[15]Pattyn, N., Van Cutsem, J., Dessy, E., & Mairesse, O. Bridging exercise science, cognitive psychology, and medical practice： Is “cognitive fatigue” a remake of “the emperor’s new clothes” ?. Frontiers in Psychology, 9 (2018), 1246. https://doi.org/10.3389/fpsyg.2018.01246.
[16]Mizuno, K., Tanaka, M., Yamaguti, K., Kajimoto, O., Kuratsune, H., & Watanabe, Y. Mental fatigue caused by prolonged cognitive load associated with sympathetic hyperactivity. Behavioral and brain functions, 7(1) (2011), 1-7. https://doi.org/10.1186/1744-9081-7-17.
[17]Chen, K., Li, Z., Ai, Q., Liu, Q., & Wang, L. An improved CNN model based on fused time-frequency features for mental fatigue detection in BCIs. In 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA) (2021, July) (pp. 1-5). IEEE. https://doi.org/10.1109/IISA52424.2021.9555518.
[18]Chai, R., Tran, Y., Naik, G. R., Nguyen, T. N., Ling, S. H., Craig, A., & Nguyen, H. T. Classification of EEG based-mental fatigue using principal component analysis and Bayesian neural network. In 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)(2016, August)(pp. 4654-4657). IEEE. https://doi.org/10.1109/EMBC.2016.7591765.
[19]Dubravac, M., & Meier, B. Overshooting cognitive control adjustments in older age： Evidence from conflict-and error-related slowing in the Stroop, Simon, and flanker tasks. Acta psychologica, 234 (2023), 103874. https://doi.org/10.1016/j.actpsy.2023.103874.
[20]Beringer, M., Wacker, J., & Recio, G. Deliberate control of facial expressions in a go/no-go task： An ERP study. Acta Psychologica, 230 (2022), 103773. https://doi.org/10.1016/j.actpsy.2022.103773.
[21]Caton, R. The electric currents of the brain. Br Med J, 2(1875), 278. https://doi.org/10.1080/00029238.1970.11080764.
[22]Gloor, P. Hans Berger on electroencephalography. American Journal of EEG Technology, 9(1) (1969), 1-8. https://doi.org/10.1080/00029238.1969.11080728.
[23]Hayward, R. The tortoise and the love-machine： Grey Walter and the politics of electroencephalography. Science in Context, 14(4) (2001), 615-641. https://doi.org/10.1017/S0269889701000278.
[24]Jeon, J., & Cai, H. Multi-class classification of construction hazards via cognitive states assessment using wearable EEG. Advanced Engineering Informatics, 53 (2022), 101646. https://doi.org/10.1016/j.aei.2022.101646.
[25]Kobayashi, K., James, C. J., Yoshinaga, H., Ohtsuka, Y., & Gotman, J. The electroencephalogram through a software microscope： non-invasive localization and visualization of epileptic seizure activity from inside the brain. Clinical neurophysiology, 111(1) (2000), 134-149. https://doi.org/10.1016/S1388-2457(99)00202-3
[26]Gu, X., Cao, Z., Jolfaei, A., Xu, P., Wu, D., Jung, T. P., & Lin, C. T. EEG-based brain-computer interfaces (BCIs)： A survey of recent studies on signal sensing technologies and computational intelligence approaches and their applications. IEEE/ACM transactions on computational biology and bioinformatics, 18(5) (2021), 1645-1666. https://doi.org/10.1109/TCBB.2021.3052811.
[27]Raghu, S., Sriraam, N., Temel, Y., Rao, S. V., Hegde, A. S., & Kubben, P. L. Performance evaluation of DWT based sigmoid entropy in time and frequency domains for automated detection of epileptic seizures using SVM classifier. Computers in biology and medicine, 110(2019), 127-143. https://doi.org/10.1016/j.compbiomed.2019.05.016.
[28]Cao, Z., Chuang, C. H., King, J. K., & Lin, C. T. Multi-channel EEG recordings during a sustained-attention driving task. Scientific Data, 6(1) (2019), 19. https://doi.org/10.1038/s41597-019-0027-4.
[29]黃浩權（2017）。基於多通道濾波共同空間模式之憂鬱症腦波分析。﹝碩士論文。國立臺北科技大學﹞臺灣博碩士論文知識加值系統。 https://hdl.handle.net/11296/9x26c7。
[30]La Rocca, D., Campisi, P., Vegso, B., Cserti, P., Kozmann, G., Babiloni, F., & Fallani, F. D. V. Human brain distinctiveness based on EEG spectral coherence connectivity. IEEE transactions on Biomedical Engineering, 61(9) (2014), 2406-2412. https://doi.org/10.1109/TBME.2014.2317881.
[31]Raghu, S., Sriraam, N., Kumar, G. P., & Hegde, A. S. A novel approach for real-time recognition of epileptic seizures using minimum variance modified fuzzy entropy. IEEE Transactions on Biomedical Engineering, 65(11) (2018), 2612-2621. https://doi.org/10.1109/TBME.2018.2810942.
[32]宋加力（2021）。區分重度憂鬱和焦慮症狀共病患者與健康對照者的腦電圖機器學習方法。﹝博士論文。國立中山大學﹞臺灣博碩士論文知識加值系統。 https://hdl.handle.net/11296/yn4dxb。
[33]Lal S. K., Craig A. A critical review of the psychophysiology of driver fatigue. Biological Psychology, 55(3) (2001), 173-194. https://doi.org/10.1016/S0301-0511(00)00085-5.
[34]Lal S. K., Craig A. Electroencephalography activity associated with driver fatigue： implications for a fatigue countermeasure device. Journal of Psychophysiology, 15(3) (2001), 183-189. https://doi.org/10.1027/0269-8803.15.3.183.
[35]Cajochen, C., Brunner, D. P., Krauchi, K., Graw, P. & Wirz-Justice, A. Power density in theta/alpha frequencies of the waking EEG progressively increases during sustained wakefulness. Sleep, 18(10) (1995), 890–894. https://doi.org/10.1093/sleep/18.10.890.
[36]Strijkstra, A. M., Beersma, D. G. M., Drayer, B., Halbesma, N., Daan, S. Subjective sleepiness correlates negatively with global alpha (8-12 Hz) and positively with central frontal theta (4-8 Hz) frequencies in the human resting awake electroencephalogram, Neuroscience Letters, 340 (1) (2003), 17-20. https://doi.org/10.1016/S0304-3940(03)00033-8
[37]Fan, X., Zhou, Q., Liu, Z., Xie, F. Electroencephalogram assessment of mental fatigue in visual search, Bio-medical materials and engineering, 26(s1) (2015), s1455–s1463. http://dx.doi.org/10.3233/BME-151444.
[38]Hu S, Zheng G, Peters B. Driver fatigue detection from electroencephalogram spectrum after electrooculography artefact removal, IET Intelligent Transport Systems, 7(1) (2013), 105-113. https://doi.org/10.1049/iet-its.2012.0045.
[39]Q. He, W. Li, X. Fan and Z. Fei. Driver fatigue evaluation model with integration of multi-indicators based on dynamic Bayesian network, IET Intell. Transp. Syst., 9(5) (2014), 547-554. https://doi.org/10.1049/iet-its.2014.0103.
[40]Craig, A., Tran, Y., Wijesuriya, N., Nguyen, H. Regional brain wave activity changes associated with fatigue. Psychophysiology, 49(4) (2012), 574–582. https://doi.org/10.1111/j.1469-8986.2011.01329.x.
[41]Hosseini, M. P., Hosseini, A., & Ahi, K. A review on machine learning for EEG signal processing in bioengineering. IEEE reviews in biomedical engineering, 14 (2020), 204-218. https://doi.org/10.1109/RBME.2020.2969915.
[42]Vivaldi, N., Caiola, M., Solarana, K., & Ye, M. Evaluating performance of eeg data-driven machine learning for traumatic brain injury classification. IEEE Transactions on Biomedical Engineering, 68(11) (2021), 3205-3216. https://doi.org/10.1109/TBME.2021.3062502.
[43]Yoo, J. On-chip epilepsy detection： Where machine learning meets patient-specific healthcare. In 2017 International SoC Design Conference (ISOCC) (pp. 146-147) (2017, November). IEEE. https://doi.org/10.1109/ISOCC.2017.8368839.
[44]Miyata, Y., Zhu, Y., Ozawa, K., Shimomura, H., & Shirasaka, T. Identification of Switches by Machine Learning of EEG Data When Listening to Switch Sounds. In 2022 IEEE 4th Global Conference on Life Sciences and Technologies (LifeTech) (pp. 219-221) (2022, March). IEEE. https://doi.org/10.1109/LifeTech53646.2022.9754762.
[45]Yang, W., Joo, M., Kim, Y., Kim, S. H., & Chung, J. M. Hybrid Machine Learning Scheme for Classification of BECTS and TLE Patients Using EEG Brain Signals. IEEE Access, 8 (2020), 218924-218935. https://doi.org/10.1109/ACCESS.2020.3038948.
[46]Murtazina, M. S., & Avdeenko, T. V. Classification of brain activity patterns using machine learning based on EEG data. In 2020 1st International Conference Problems of Informatics, Electronics, and Radio Engineering (PIERE) (pp. 219-224) (2020, December). IEEE. https://doi.org/10.1109/PIERE51041.2020.9314660.
[47]Borragán, G., Slama, H., Bartolomei, M., & Peigneux, P. Cognitive fatigue： A time-based resource-sharing account. Cortex, 89 (2017), 71-84. https://doi.org/10.1016/j.cortex.2017.01.023.
[48]Stone, J. V. Independent component analysis： an introduction. Trends in cognitive sciences, 6(2) (2002), 59-64. https://doi.org/10.1016/S1364-6613(00)01813-1.
[49]Saby, J. N., & Marshall, P. J. The utility of EEG band power analysis in the study of infancy and early childhood. Developmental neuropsychology, 37(3) (2012), 253-273. https://doi.org/10.1080/87565641.2011.614663.
[50]Cajochen, C., Brunner, D. P., Krauchi, K., Graw, P., & Wirz-Justice, A. Power density in theta/alpha frequencies of the waking EEG progressively increases during sustained wakefulness. Sleep, 18(10) (1995), 890-894. https://doi.org/10.1093/sleep/18.10.890.
[51]Vinck, M., Oostenveld, R., Van Wingerden, M., Battaglia, F., & Pennartz, C. M. An improved index of phase-synchronization for electrophysiological data in the presence of volume-conduction, noise and sample-size bias. Neuroimage, 55(4) (2011), 1548-1565. https://doi.org/10.1016/j.neuroimage.2011.01.055.
[52]Li, H., Liu, T., Wu, X., & Chen, Q. A bearing fault diagnosis method based on enhanced singular value decomposition. IEEE Transactions on Industrial Informatics, 17(5) (2020), 3220-3230. https://doi.org/10.1109/TII.2020.3001376.
[53]Henry, E. R., & Hofrichter, J. Singular value decomposition： Application to analysis of experimental data. In Methods in enzymology (Vol. 210, pp. 129-192) (1992). Academic Press. https://doi.org/10.1016/0076-6879(92)10010-B.
[54]Golafshan, R., & Sanliturk, K. Y. SVD and Hankel matrix based de-noising approach for ball bearing fault detection and its assessment using artificial faults. Mechanical Systems and Signal Processing, 70 (2016), 36-50. https://doi.org/10.1016/j.ymssp.2015.08.012.
[55]De Vries, J., Michielsen, H., Van Heck, G. L., & Drent, M. Measuring fatigue in sarcoidosis： the Fatigue Assessment Scale (FAS). British journal of health psychology, 9(3) (2004), 279-291. https://doi.org/10.1348/1359107041557048.
[56]Xu, T., Xu, L., Zhang, H., Ji, Z., Li, J., Bezerianos, A., & Wang, H. Effects of Rest-Break on mental fatigue recovery based on EEG dynamic functional connectivity. Biomedical Signal Processing and Control, 77(2022), 103806. https://doi.org/10.1016/j.bspc.2022.103806.
[57]Plechawska-Wojcik, M., Kaczorowska, M., & Zapala, D. The artifact subspace reconstruction (ASR) for EEG signal correction. A comparative study. In Information systems architecture and technology: proceedings of 39th international conference on information systems architecture and technology–ISAT 2018: part II (pp. 125-135) (2019). Springer International Publishing. https://doi.org/10.1007/978-3-319-99996-8_12.
[58]Li, S., Zhou, W., Yuan, Q., Geng, S., & Cai, D. Feature extraction and recognition of ictal EEG using EMD and SVM. Computers in biology and medicine, 43(7)(2013), 807-816. https://doi.org/10.1016/j.compbiomed.2013.04.002.
[59]Haresign, I. M., Phillips, E., Whitehorn, M., Noreika, V., Jones, E. J. H., Leong, V., & Wass, S. V. Automatic classification of ICA components from infant EEG using MARA. Developmental cognitive neuroscience, 52(2021), 101024. https://doi.org/10.1016/j.dcn.2021.101024.
[60]Chuang, C. H., Chang, K. Y., Huang, C. S., & Jung, T. P. IC-U-Net: a U-Net-based denoising autoencoder using mixtures of independent components for automatic EEG artifact removal. NeuroImage, 263(2022), 119586. https://doi.org/10.1016/j.neuroimage.2022.119586.
[61]Jap, B. T., Lal, S., Fischer, P., & Bekiaris, E. Using EEG spectral components to assess algorithms for detecting fatigue. Expert Systems with Applications, 36(2) (2009), 2352-2359. https://doi.org/10.1016/j.eswa.2007.12.043.
[62]Ten Caat, M., Lorist, M. M., Bezdan, E., Roerdink, J. B., & Maurits, N. M. High-density EEG coherence analysis using functional units applied to mental fatigue. Journal of neuroscience methods, 171(2) (2008), 271-278. https://doi.org/10.1016/j.jneumeth.2008.03.022.
[63]Xu, T., Xu, L., Zhang, H., Ji, Z., Li, J., Bezerianos, A., & Wang, H. Effects of Rest-Break on mental fatigue recovery based on EEG dynamic functional connectivity. Biomedical Signal Processing and Control, 77(2022), 103806. https://doi.org/10.1016/j.bspc.2022.103806
[64]Smith, S. M. (2004). Overview of fMRI analysis. The British Journal of Radiology, 77(suppl_2), S167-S175. https://doi.org/10.1259/bjr/33553595.
[65]Tai, Y. F., & Piccini, P. (2004). Applications of positron emission tomography (PET) in neurology. Journal of Neurology, Neurosurgery & Psychiatry, 75(5), 669-676. https://doi.org/10.1136/jnnp.2003.028175.
[66]Buzug, T. M. (2011). Computed tomography. In Springer handbook of medical technology (pp. 311-342). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-74658-4_16.
[67]Sun, Y., Lim, J., Kwok, K., & Bezerianos, A. (2014). Functional cortical connectivity analysis of mental fatigue unmasks hemispheric asymmetry and changes in small-world networks. Brain and cognition, 85, 220-230. https://doi.org/10.1016/j.bandc.2013.12.011.