基於閾值搜索最大相關最小冗餘演算法與腦電圖訊號分析之睡眠呼吸中止事件分類

簡易檢索 / 詳目顯示

回結果列表

研究生：	謝佳彤 Hsieh, Chia-Tung
論文名稱：	基於閾值搜索最大相關最小冗餘演算法與腦電圖訊號分析之睡眠呼吸中止事件分類 Classification of Sleep Apnea Events Using Threshold Search Maximum Relevance Minimal Redundancy Algorithm with EEG Signals
指導教授：	黃元豪 Huang, Yuan-Hao
口試委員:	劉奕汶 Liu, Yi-Wen 李祈均 Lee, Chi-Chun 黃柏鈞 Huang, Po-Chiun 范倫達 Van, Lan-Da
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2025
畢業學年度：	113
語文別：	英文
論文頁數：	75
中文關鍵詞：	睡眠呼吸中止症、腦電圖訊號、mRMR特徵選擇、TSmRMR特徵選擇、機器學習
外文關鍵詞：	Sleep Apnea-Hypopnea Syndrome, EEG Signals, mRMR Feature Selection, TSmRMR Feature Selection, Machine Learning
相關次數：	點閱：64 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

睡眠呼吸中止症（Sleep Apnea）是一種在睡眠期間呼吸頻繁中斷，容易對生活品質造成嚴重影響的疾病。這些呼吸中斷時間從數秒至一分鐘以上，會干擾深層睡眠，導致白天嗜睡、認知障礙，並增加意外風險。在傳統上，利用多導睡眠監測（Polysomnography, PSG）記錄呼吸、血氧飽和度和腦部活動等生理訊號並進行診斷。然而，這項檢測需要將大量的感測器放置在受測者身上，可能引起不適並影響自然睡眠模式。為了降低病患對感測器的不適感，本研究利用腦電圖訊號（Electroencephalography, EEG）作為替代方法。腦電圖通過捕捉睡眠期間的詳細腦部活動，可在不依賴多導睡眠監測的情況下準確檢測睡眠呼吸中止症事件發生。由於資料的複雜性，在提取與選擇腦電圖訊號的特徵時面臨挑戰。為了解決此問題，本研究提出了一種特徵選擇方法，稱為「閾值搜索最大相關最小冗餘演算法」（Threshold Search Maximal Relevance Minimal Redundancy, TSmRMR）。此方法在傳統mRMR特徵選擇方法中加入閾值搜索方法，從腦電圖訊號的時、頻域特徵中選取更具消息性且冗餘度更低者，從而提升分類準確率。

本研究使用了兩個開放資料集。其中一個資料集提供了較全面且較符合實際發生情況的呼吸中止事件，另一個資料集包含的事件記錄較少，資料量有限。研究結果顯示，使用K近鄰（KNN）分類器時，TSmRMR演算法在兩個資料集上的分類準確率分別為：使用29個特徵達到96.97% 和使用單一特徵達到98.90%。此外，TSmRMR演算法能夠有效應對醫療數據中常見的資料不平衡問題。本研究不僅提升了分類結果，還能在不同資料集中保持穩定的準確性，是一種可靠的基於腦電圖訊號的睡眠呼吸中止事件診斷方法。

Sleep apnea is characterized by frequent interruptions in breathing during sleep, significantly impairing the quality of life. These interruptions, lasting from a few seconds to over a minute, disrupt restful sleep and lead to excessive daytime drowsiness, cognitive impairments, and an increased risk of accidents. The standard diagnostic method, polysomnography (PSG), monitors physiological signals, including respiratory effort, oxygen saturation, and brain activity. However, PSG involves the placement of numerous sensors on the body, causing discomfort and potentially disrupting natural sleep patterns. This study uses electroencephalography (EEG) signals as an alternative method to reduce patient discomfort. By capturing brain activity during sleep, EEG allows for reliable detection of sleep apnea without requiring the full PSG test. Extracting and selecting features from EEG signals presents challenges due to the complexity of the data. To address this issue, this study introduces a novel feature selection method called threshold search maximal relevance minimal redundancy (TSmRMR). By incorporating a threshold search method into the traditional maximal relevance minimal redundancy (mRMR) feature selection method, this approach selects more informative and less redundant features from the time-frequency domain of EEG signals, enhancing classification accuracy.

This study uses two publicly available datasets. One provides a comprehensive and realistic representation of apnea events, while the other offers fewer recorded events. The proposed method achieves classification accuracies of 96.97\% using 29 features and 98.90\% using a single feature with the K-nearest neighbor (KNN) classifier across both datasets. Additionally, TSmRMR effectively addresses data imbalance, a common challenge in medical datasets. These findings demonstrate that the proposed method enhances classification performance while ensuring consistent accuracy across diverse dataset conditions, establishing it as a reliable method for real-world sleep apnea diagnosis using EEG signals.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1 Sleep Apnea-hypopnea Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 Sleep Apnea Event Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.1 Obstructive Sleep Apnea (OSA) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 Central Sleep Apnea (CSA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.3 Hypopnea (HYP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3 Research Motivation . . . . . . . . . .  . . . . . . . . . . . . . . . . . . . . . . . 5
4 Related Works . . . . . . . . . . . . . .  . . . . . . . . . . . . . . . . . . . . . . 7
Medical Diagnosis of Sleep Apnea . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1 Polysomnography (PSG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 Electroencephalography (EEG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.1 Electrode Placement in EEG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Sub-band Decomposition of EEG . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Correlation Between EEG Activity and Sleep Apnea. . . . . . . . . . . . . . . . . . 13
Proposed Sleep Apnea Event Classification Algorithm . . . . . . . . . . . . . . . . . . 19
1 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2 Signal Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.1 Filtering and Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.1 Signal Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3 Feature Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 27
3.1 Time Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2 Frequency Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3 Cascading Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4 Feature Selection Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1 Mutual Information (MI). . . . . . . . . . . . . . . . . . .  . . . . . . . . . . . 32
4.2 Maximal Relevance Minimal Redundancy (mRMR) . . . . . . . . . . . . . . . . . . . . 36
4.3 Proposed Threshold Search mRMR (TSmRMR) . . . . . . . . . . . . . . . . . . . . . . 39
5 K-nearest Neighbors (KNN) Classifier . . . . . . . . . . . . . . . . . . . . . . . . . 43
Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
1 Simulation Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2 Feature Selection Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.1 Simulation of the MIT-BIH Dataset . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.2 Simulation of the UCDDB Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.3 Performance Analysis for Feature Selection Methods. . . . . . . . . . . . . . . . . . 56
2.4 Performance Analysis for Sub-band EEG Signals . . . . . . . . . . . . . . . . . . . . 58
3 Classification Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.1 Performance Analysis for Classification . . . . . . . . . . . . . . . . . . . . . . . 59
3.2 Performance Analysis with Previous Studies . . . . . . . . . . . . . . . . . . . . . 66
Conclusion and Future Works. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
2 Future Works. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

                                

[1] Mit-bih polysomnographic database. Available online at: https://physionet.org/content/slpdb/1.0.0/, accessed on 2024-10-15.
[2] St. vincent’s university hospital / university college dublin sleep apnea database,
18.02.2011. Available online at: https://physionet.org/content/ucddb/1.0.0/, accessed on 2024-10-15.
[3] Jhao-Cheng Wu and Yuan-Hao Huang. Sleep apnea syndrome screening by tri-axial accelerometer, oximeter and phenotype information. Master’s thesis, National
Tsing Hua University, 2017.
[4] Philip L Smith, Robert A Wise, Avram R Gold, Alan R Schwartz, and Solbert Permutt. Upper airway pressure-flow relationships in obstructive sleep apnea. Journal of applied physiology, 64(2):789–795, 1988.
[5] Paul M Macey, Rajesh Kumar, Mary A Woo, Edwin M Valladares, Frisca L Yan-Go, and Ronald M Harper. Brain structural changes in obstructive sleep apnea. Sleep, 31(7):967–977, 2008.
[6] Shahrokh Javaheri, Ferran Barbe, Francisco Campos-Rodriguez, Jerome A Dempsey, Rami Khayat, Sogol Javaheri, Atul Malhotra, Miguel A Martinez-Garcia, Reena Mehra, Allan I Pack, et al. Sleep apnea: types, mechanisms, and clinical cardiovascular consequences. Journal of the American College of Cardiology, 69(7):841–858, 2017.
[7] Ian G Campbell. Eeg recording and analysis for sleep research. Current protocols in neuroscience, 49(1):10–2, 2009.
[8] Richard B Berry, Rita Brooks, Charlene E Gamaldo, Susan M Harding, Carole Marcus, Bradley V Vaughn, et al. The aasm manual for the scoring of sleep and associated events. Rules, Terminology and Technical Specifications, Darien, Illinois, American Academy of Sleep Medicine, 176(2012):7, 2012.
[9] MD Lewis R Kline. Clinical presentation and diagnosis of obstructive sleep apnea in adults. Available online at: https://www.uptodate.com/contents/clinical-presentation-and-diagnosis-of-obstructive-sleep-apnea-in-adults, accessed on 2024-11-08.
[10] MD M Safwan Badr. Central sleep apnea: risk factors, clinical presentation, and diagnosis. Available online at: https://www.uptodate.com/contents/central-sleep-apnea-risk-factors-clinical-presentation-and-diagnosis, accessed on 2024-11-08.
[11] Arnab Bhattacharjee, Suvasish Saha, Shaikh Anowarul Fattah, Wei-Ping Zhu, and M Omair Ahmad. Sleep apnea detection based on rician modeling of feature variation in multiband eeg signal. IEEE journal of biomedical and health informatics, 23(3):1066–1074, 2018.
[12] Xiaoyun Zhao, Xiaohong Wang, Tianshun Yang, Siyu Ji, Huiquan Wang, Jinhai Wang, Yao Wang, and Qi Wu. Classification of sleep apnea based on eeg sub-ban signal characteristics. Scientific Reports, 11(1):5824, 2021.
[13] Arnab Chatterjee and Nanda Dulal Jana. Classification of sleep apnea event type using imbalanced labelled eeg signal. In 2022 IEEE Region 10 Symposium (TENSYMP), pages 1–6, 2022.
[14] Atiya Khan, Saroj Kr. Biswas, Chukhu Chunka, and Akhil Kumar Das. A machine learning model for obstructive sleep apnea detection using ensemble learning and single-lead eeg signal data. IEEE Sensors Journal, 24(12):20266–20273, 2024.
[15] Behnam Gholami, Mohammad Hossein Behboudi, Mohammad Ghasem Mahjani, and Ali Khadem. Diagnosis of sleep apnea syndrome from eeg signals using different entropy measures. In 2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS), pages 1–6, 2021.
[16] Saba Bayatfar, Saman Seifpour, Mohammadreza Asghari Oskoei, and Ali Khadem. An automated system for diagnosis of sleep apnea syndrome using single-channel eeg signal. In 2019 27th Iranian Conference on Electrical Engineering (ICEE), pages 1829–1833, 2019.
[17] Edward A Wolpert. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Archives of General Psychiatry, 20(2):246–247, 1969.
[18] Torbjorn ˚Akerstedt and Peter M Nilsson. Sleep as restitution: an introduction. Journal of internal medicine, 254(1):6–12, 2003.
[19] Sudhansu Chokroverty. Overview of sleep & sleep disorders. Indian Journal of Medical Research, 131(2):126–140, 2010.
[20] Jessica Vensel Rundo and Ralph Downey III. Polysomnography. Handbook of clinical neurology, 160:381–392, 2019.
[21] Michal Teplan et al. Fundamentals of eeg measurement. Measurement science
review, 2(2):1–11, 2002.
[22] Priyanka A. Abhang, Bharti W. Gawali, and Suresh C. Mehrotra. Chapter 2-technological basics of eeg recording and operation of apparatus. In Priyanka A. Abhang, Bharti W. Gawali, and Suresh C. Mehrotra, editors, Introduction to EEG and Speech-Based Emotion Recognition, pages 19–50. Academic Press, 2016.
[23] E.J. Smith. Introduction to eeg. Available online at: https://www.ebme.co.uk/articles/clinical-engineering/introduction-to-eeg, accessed on 2024-11-08.
[24] A. L. Goldberger, L. A. N. Amaral, L. Glass, J. M. Hausdorff, P. Ch. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, and H. E. Stanley. PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation, 101(23):e215–e220, 2000 (June 13). Circulation Electronic Pages: http://circ.ahajournals.org/content/101/23/e215.full PMID:1085218; doi: 10.1161/01.CIR.101.23.e215.
[25] Thomas M Cover, Joy A Thomas, et al. Entropy, relative entropy and mutual
information. Elements of information theory, 2(1):12–13, 1991.
[26] David Freedman and Persi Diaconis. On the histogram as a density estimator: L 2 theory. Zeitschrift fur Wahrscheinlichkeitstheorie und verwandte Gebiete, 57(4):453 476, 1981.
[27] P. Welch. The use of fast fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Transactions on Audio and Electroacoustics, 15(2):70–73, 1967.
[28] Kendall Atkinson. An introduction to numerical analysis. John wiley & sons, 1991.
[29] Hanchuan Peng, Fuhui Long, and Chris Ding. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Transactions on pattern analysis and machine intelligence, 27(8):1226–1238, 2005.
[30] C. Ding and H. Peng. Minimum redundancy feature selection from microarray gene expression data. In Computational Systems Bioinformatics. CSB2003. Proceedings of the 2003 IEEE Bioinformatics Conference. CSB2003, pages 523–528, 2003.
[31] Evelyn Fix and Joseph Lawson Hodges. Discriminatory analysis. Nonparametric discrimination: Consistency properties. International Statistical Review/Revue Internationale de Statistique, 57(3):238–247, 1989.
[32] Naomi S Altman. An introduction to kernel and nearest-neighbor nonparametric regression. The American Statistician, 46(3):175–185, 1992.
[33] Stephen V Stehman. Selecting and interpreting measures of thematic classification accuracy. Remote sensing of Environment, 62(1):77–89, 1997.
[34] David MW Powers. Evaluation: from precision, recall and f-measure to roc, informedness, markedness and correlation. arXiv preprint arXiv:2010.16061, 2020.
[35] Jerome A Dempsey, Sigrid C Veasey, Barbara J Morgan, and Christopher P O’Donnell. Pathophysiology of sleep apnea. Physiological reviews, 90(1):47–112, 2010.
[36] Terry Young, James Skatrud, and Paul E Peppard. Risk factors for obstructive sleep apnea in adults. Jama, 291(16):2013–2016, 2004.
[37] Yuhei Ichimaru and GB Moody. Development of the polysomnographic database on cd-rom. Psychiatry and clinical neurosciences, 53(2):175–177, 1999.
[38] Leo Breiman. Random forests. Machine learning, 45:5–32, 2001.
[39] Chris Seiffert, Taghi M Khoshgoftaar, Jason Van Hulse, and Amri Napolitano. Rusboost: A hybrid approach to alleviating class imbalance. IEEE transactions on systems, man, and cybernetics-part A: systems and humans, 40(1):185–197, 2009.
[40] Nicola Michielli, U Rajendra Acharya, and Filippo Molinari. Cascaded lstm recurrent neural network for automated sleep stage classification using single-channel eeg signals. Computers in biology and medicine, 106:71–81, 2019.

簡易檢索 / 詳目顯示

相關論文