跳到主要內容

臺灣博碩士論文加值系統

(44.213.60.33) 您好!臺灣時間:2024/07/20 06:51
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:劉翰錡
研究生(外文):LIU, HAN-CHI
論文名稱:以機器學習應用於前額血氧飽和濃度之動態測量
論文名稱(外文):Dynamic Measurement of Forehead Peripheral Arterial Oxygen Saturation (SpO2) Based on Machine Learning
指導教授:譚旦旭譚旦旭引用關係劉省宏劉省宏引用關係
指導教授(外文):TAN, TAN-HSULIU, SHING-HONG
口試委員:譚旦旭劉省宏黃永發曾德樟
口試委員(外文):TAN, TAN-HSULIU, SHING-HONGHUANG, YUNG FATSENG, DER-CHANG
口試日期:2020-07-08
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:電機工程系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:84
中文關鍵詞:血氧飽和濃度PPG訊號脈搏波波形支援向量機
外文關鍵詞:Oxygen saturation (SpO2)Photoplethysmography (PPG)Pulse waveSupport Vector Machine (SVM)
相關次數:
  • 被引用被引用:1
  • 點閱點閱:773
  • 評分評分:
  • 下載下載:53
  • 收藏至我的研究室書目清單書目收藏:1
血氧飽和濃度(Oxygen Saturation, SpO2)的應用包含了醫療、運動休閒以及睡眠品質檢測等,是評估身體健康狀況的重要參數。目前市面上常見的脈搏血氧儀多為指夾式,其量測期間容易受到移動干擾及環境光等因素的影響,造成數值失準,同時也影響到受測者的活動性。本研究旨在製作一套前額式脈搏血氧濃度儀,用於測量人體的前額PPG訊號,並提高動態活動時SpO2的準確度。本系統採用MAX30102晶片組成陣列式感測器,可以在量測時選擇PPG訊號較佳的位置進行量測。為檢測出受干擾之PPG訊號,本研究依據SpO2的數值、相對誤差及脈搏波之形態分析,作為標記區段PPG訊號品質的標準,最後分別從紅光(Red)及紅外光(IR)PPG訊號中,提取出12種特徵輸入至SVM進行PPG訊號品質的分類。
本研究共邀請20位受測者參與實驗,並分別進行靜態和動態的測量。經過SVM分類的結果顯示其平均準確率(Accuracy)、特異性(Specificity)、靈敏度(Sensitivity)及精確度(Precision)分別為89.90%、87.27%、91.82%及90.77%,並且能有效地降低SpO2的誤差及標準差,故可增進脈搏血氧儀在動態測量時的準確度,進一步提升本研究的實用性。

The applications of Oxygen Saturation (SpO2) include medical treatment, sports, leisure and sleep monitoring. It plays an importance role in assessing physical heath. At present, most of the pulse oximeters on the market are finger-type which are easily affected by motion artifact and ambient light during the measurement. These corruptions reduce SpO2 accuracy, and this sensor probe also limits the user’s mobility. In this study, a forehead pulse oximeter which measured forehead Photoplethysmographic (PPG) signal with a higher accuracy for the SpO2 measurement is developed. This system uses the integrated chip (MAX30102) to build an array-type multi-sensor probe, which is able to select a better position for measuring the SpO2. In order to detect the corrupted PPG signal, we use SpO2 value, relative error and PPG wave morphology as the standard for labeling the quality of PPG signal. Finally, 12 features are extracted separately from red and infrared (IR) PPG signals for training and testing trials. Then the SVM is employed to classify the quality of PPG signal.
We have invited 20 volunteers to participate in this study. In the experiment, a series of static and dynamic measurements are performed. The results of SVM-based classification show that the average accuracy, specificity, sensitivity and precision are respectively 89.90%, 87.27%, 91.82%, and 90.77%. Moreover, the error and standard deviation of SpO2 have been reduced effectively, which indicate that the accuracy of SpO2 during dynamic measurement can be improved, thus enhancing the practicality of this work.

摘 要 i
ABSTRACT iii
誌 謝 v
目 錄 vi
表目錄 ix
圖目錄 x
第一章 緒論 1
1.1 研究背景與動機 1
1.2 文獻探討 2
1.3 研究方法與目的 4
1.4 論文架構 4
第二章 背景知識 6
2.1 光體積變化描記圖 6
2.1.1 測量原理 6
2.1.2 脈搏波 7
2.1.3 PPG感測器 8
2.2 血氧飽和濃度 9
2.2.1 定義 9
2.2.2 量測及估算方式 9
2.3 影響訊號品質的因素 11
2.3.1 線性支持向量機 14
2.3.2 非線性支持向量機 16
第三章 前額式脈搏血氧量測系統 18
3.1 系統架構 18
3.2 硬體元件 19
3.2.1 微控制器 19
3.2.2 脈搏血氧及心律感測器 20
3.2.3 三軸加速度計 21
3.2.4 藍芽模組 21
3.3 硬體架構及實體 22
3.4 軟、韌體架構 23
3.4.1 MCU主程式執行流程 23
3.4.2 MCU中斷副程式流程 24
3.5 數位訊號處理之執行流程 26
3.5.1 訊號反向(Signal Inversion) 27
3.5.2 低通濾波器(Low-pass filter) 27
3.5.3 高通濾波器(High-pass filter) 28
3.6 圖形化使用者介面 30
第四章 資料分析 31
4.1 波形分割及SpO2之估算 31
4.1.1 脈搏波分割(Pulse Wave Segment) 31
4.1.2 血氧飽和濃度(SpO2)的估算 34
4.2 PPG訊號之標記 36
4.2.1 脈搏波之形態特徵 36
4.2.2 脈搏波品質檢測 38
4.2.3 PPG區段訊號品質標記方法 42
4.3 PPG訊號之特徵參數選取 44
第五章 實驗方法 48
5.1 測量方法 48
5.1.1 測量裝置 48
5.1.2 最佳位置測量 49
5.1.3 靜態測量 52
5.1.4 動態測量 53
5.2 SVM分類器 55
5.3 評估方法 57
第六章 實驗結果 59
6.1 資料數量及標記結果 59
6.2 實驗一:最佳參數找尋 60
6.3 實驗二:SVM效能評估 62
6.3.1 分類結果之效能 62
6.3.2 SpO2之變化 64
6.4 實驗三:SVM以靜態資料估測動態資料 72
第七章 討論與結論 74
7.1 SVM分類結果之討論 74
7.2 結論 77
7.3 未來展望 78
參考文獻 79


[1]K. H. Shelley, “Photoplethysmography: beyond the calculation of arterial oxygen saturation and heart rate,” Anesthesia and analgesia, vol. 105, no. 6 Supplement, pp. S31–6, 2007.
[2]T. Aoyagi and K. Miyasaka, “Pulse oximetry: its invention, contribution to medicine, and future tasks,” Anesthesia and analgesia, vol. 94,1, 2002.
[3]M. Ascha, A. Bhattacharyya, J.A. Ramos, and A.R. Tonelli, “Pulse oximetry and arterial oxygen saturation during cardiopulmonary exercise testing,” Medicine & Science in Sports & Exercise, pp. 1992-1997, 2018.
[4]王筱涵,以血氧飽和濃度檢測睡眠呼吸中止症,碩士論文,國立中山大學機械與機電工程學系,2006
[5]T. Tamura, “Current progress of photoplethysmography and SPO2 for health monitoring,” Biomedical Engineering. Letters, vol. 9, no. 1, pp. 21-36, Feb. 2019.
[6]J. Wannenburg and R. Malekian, “Body sensor network for mobile health monitoring, a diagnosis and anticipating system,” IEEE Sensors Journal, vol. 15, no. 12, pp. 6839-6852, 2015.
[7]A. Bideaux, T.Schwendemann, and S. Hey, “Evaluation of design parameters for a reflection based long-term pulse oximetry sensor,” Proceedings of the 5th EAI International Conference on Wireless Mobile Communication and Healthcare, pp 190-193, 2015.
[8]M. T. Petterson, V. L. Begnoche, and J. M. Graybeal, “The effect of motion on pulse oximetry and its clinical significance,” Anesthesia and analgesia, vol. 105, no. 6, pp. 78–84, 2007.
[9]Sugino S, Kanaya N, Mizuuchi M, Nakayama M, and Namiki A, “Forehead is as sensitive as finger pulse oximetry during general anesthesia,” Canadian Journal of Anesthesia, 51(5):432-6, 2004.
[10]D.E. Bebout, P.D. Mannheimer, and C. C. Wun, “Site-dependent differences in the time to detect changes in saturation during low perfusion,” Critical Care Medicine, vol. 29, no. 12, 2001.
[11]V. Blaylock, et al., “Comparison of finger and forehead oximetry sensors in postanesthesia care patients,” Journal of PeriAnesthesia Nursing, vol. 23, No. 6, pp. 379-386, 2008.
[12]Z. Lu, X. Chen, Z. Dong, Z. Zhao, and X. Zhang, “A prototype of reflection pulse oximeter designed for mobile healthcare,” IEEE Journal of Biomedical and Health Informatics, vol. 20, no. 5, pp. 1309-1320, 2016.
[13]J. Wan, Y. Zou, Y. Li, and J. Wang, “Reflective type blood oxygen saturation detection system based on MAX30100,” 2017 International Conference on Security, Pattern Analysis, and Cybernetics, pp. 615-619, 2017.
[14]A. Azhari, et al., “A patch-type wireless forehead pulse oximeter for SpO2 measurement,” 2017 IEEE Biomedical Circuits and Systems Conference, pp. 1-4, 2017.
[15]B. S. Kim and S. K. Yoo, “Motion artifact reduction in photoplethysmography using independent component analysis,” IEEE Transactions on Biomedical Engineering, vol. 53, no. 3, pp. 566-568, 2006.
[16]M. R. Ram, K. V. Madhav, E. H. Krishna, N. R. Komalla, and K. A. Reddy, “A novel approach for motion artifact reduction in PPG signals based on AS-LMS adaptive filter,” IEEE Transactions on Instrumentation and Measurement, vol. 61, no. 5, pp. 1445-1457, 2012.
[17]T. L. Rusch, R. Sankar, and J. E. Scharf, “Signal processing methods for pulse oximetry,” Computers in Biology and Medicine, vol. 26, no. 2, pp. 143–159, 1996.
[18]C. Yu, Z. Liu, T. McKenna, A. T. Reisner, and J. Reifman, “A method for automatic identification of reliable heart rates calculated from ECG and PPG waveforms,” Journal of the American Medical Informatics Association, vol. 13, no. 3, pp. 309–320, 2006.
[19]P. Gibbs and H. H. Asada, “Reducing motion artifact in wearable biosensors using MEMS accelerometers for active noise cancellation,” 2005 American Control Conference, vol. 3, pp. 1581–1586, 2005.
[20]J. A. Sukor, S. J. Redmond, and N. H. Lovell, “Signal quality measures for pulse oximetry through waveform morphology analysis,” Physiological Measurement, vol. 32, no. 3, pp. 369–384, 2011.
[21]C. Fischer, B. Dömer, T. Wibmer, and T. Penzel, “An algorithm for real-time pulse waveform segmentation and artifact detection in photoplethysmograms,”z IEEE Journal of Biomedical and Health Informatics, vol. 21, no. 2, pp. 372-381, 2017.
[22]A. C. Burton, “The range and variability of the blood flow in the human fingers and the vasomotor regulation by body temperature,” American Physiological Society, vol.127, pp.437-453, 1939.
[23]M. Nitzan, A. Babchenko, D. Shemesh, and J. Alberton, “Influence of thoracic sym- pathectomy on cardiac induced oscillations in tissue blood volume,” Medical & Biological Engineering & Computing, vol. 39, pp.579-583, 2001.
[24]S. Bagha and L. Shaw, “A real time analysis of PPG signal for measurement of SpO2 and pulse rate,” International Journal of Computer Applications, vol. 36, no. 11, pp. 45–50, 2011.
[25]S. Fouzas, K. N. Priftis, and M. B. Anthrcopoulos. “Pulse oximetry in pediatric practice,” Pediatrics, 128(4):740–752, 2011.
[26]林育德,脈波信號與PPG信號特徵之相關性研究,中醫藥年報,第28期,第一冊,2010。
[27]林豫亭,探討小兒加護病房病童之SpO2與SaO2的關係,碩士論文,中山醫學大學醫學研究所,臺中,2003。
[28]李仁貴,穿戴式生醫量測原理與實驗:使用Android Studio開發,全華科技圖書公司,台灣,台北,03,2017,第213-219頁。
[29]K. A. Reddy, B. George, N. M. Mohan, and V. J. Kumar, “A novel calibration-free method of measurement of oxygen saturation in arterial blood,” IEEE Transaction on Instrumentation Measurement, vol. 58, no. 5, pp. 1699–1705, 2009.
[30]M. Tian, W. Liu, Z. Zhong, and Y. Pan, “Design and research on the hardware of wireless oxygen saturation detection based on ARM,” International Conference on Intelligent Control and Information Processing, pp. 707–710, 2010.
[31]J. W. Chong, D. K. Dao, S. M. A. Salehizadeh, D. D. McManus, C. E. Darling, K. H. Chon, and Y. Mendelson, “Photoplethysmograph signal reconstruction based on a novel hybrid motion artifact detection-reduction approach. Part I: Motion and noise artifact detection,” Annals of Biomedical Engineering, vol. 42, no. 11, pp. 2238–2250, 2014.
[32]R. M. Tobin, J. A. Pologe, and P. B. Batchelder, “A characterization of motion affecting pulse oximetry in 350 patients,” Anesthesia and Analgesia, vol. 94, no. 1, pp. S54–61, 2002.
[33]C. Cortes and V. Vapnik, “Support-vector network,” Machine Learning, 20:273–297, 1995.
[34]C. J. C. Burges. “A tutorial on support vector machines for pattern recognition,” Data Mining and Knowledge Discovery, Vol.2, pp. 955-974, 1998.
[35]林宗勳,Support Vector Machine 簡介,資料來源:http://www.cmlab.csie.ntu.edu.tw/~cyy/learning/tutorials/SVM2.pdf.
[36]孫聖育,以叢集為基礎的支撐向量機學習及其應用於語者辨識,碩士論文,國立交通大學,資訊工程學系,第3-9頁,2004。
[37]陳盈秀,SVM類神經網路於單調性資料探勘之研究,碩士論文,國立成功大學,工業與資訊管理學系,第24-27頁,2009。
[38]C. W. Hsu, C. C. Chang, and C. J. Lin, “A practical guide to support vector classification,” 2003. [online] Available: https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf.
[39]“MSP430F543xA, MSP430F541xA Mixed-Signal Micro-controllers datasheet (Rev. F),” Texas Instrument, Corp, 2014.
[40]“MAX30102 High-Sensitivity Pulse Oximeter and Heart-Rate Sensor for Wearable Health,” Maxim Integrated Products, Inc, 2018.
[41]“ADXL325: Small, Low Power, 3-Axis ±5 g Accelerometer Data Sheet (Rev. 0),” Analog Device, Inc, 2009.
[42]J. M. Cho, Y. K. Sung, and K. W. Shin, “A preliminary study on photoplethysmogram(PPG) signal analysis for reduction of motion artifact in frequency domain,” 2012 IEEE EMBS Conference on Biomedical Engineering and Sciences, pp. 28–33, 2012.
[43]S. Kimura and M. Okada, “Estimation of arterial pulse wave velocities in the frequency domain: Method and clinical considerations,” Medical and Biological Engineering and Computing, vol. 24, no. 3, pp. 255–260, 1986.
[44]Z. Ai-Hua, W. Ping, and C. Yong-Xin . “Peak detection of pulse signal based on dynamic difference threshold,” Journal of Jilin University(Engineering and Technology Edition), pp. 847-853, 2014.
[45]A. M.Weissler, W. S. Harris, and C. D. Schoenfeld, “Systolic time intervals in heart failure in man,” Circulation, vol. 37, no. 2, pp. 149–159, 1968.
[46]R. P. Lewis, S. E. Rittogers, W. F. Froester, and H. Boudoulas, “A critical review of the systolic time intervals,” Circulation, vol. 56, no. 2, pp. 146–158, 1977.
[47]Maxim Integrated Products, Inc, “Recommended configurations and operating profiles for MAX30101/MAX30102 EV Kits.” pp. 6, 2018.
[48]瑞盛醫學科技股份有限公司,SA300,SA310,SA320插電式Pulse Oximeter使用者說明書,臺北,2014。
[49]M. W. Wukitsch, M. T. Petterson, D. R. Tobler, and J. A. Pologe, “Pulse oximetry: analysis of theory, technology, and practice,” Journal of Clinical Monitoring, 4(4), 290-301, 1988.
[50]Medtronic, “NellcorTM SpO2 Forehead Sensor Brochure: Learn more about the sensor’s real-time, accurate data,” Medtronic plc, pp. 4, 2016.
[51]C.-C. Chang and C.-J. Lin. “LIBSVM: A library for support vector machines,” [online] Available: https://www.csie.ntu.edu.tw/~cjlin/libsvm/.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top