Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/98246
標題: 基於機器學習技術於非侵入式量測系統 實作探討以學習型無創血糖儀為例
Based on machine learning technology in non-invasive measurement system-A Case Study of the learning non-invasive blood glucose meter
作者: 徐正易
Cheng-Yi Hsu
關鍵字: 嵌入式系統
非侵入式血糖
迴歸分析
SVM支援向量機
Embedded system
non-invasive blood glucose
regression analysis
SVM
引用: 100 days of algorithms. (無日期). 擷取自 github: https://github.com/coells/100days 2. (2015). 擷取自 衛生福利部國民健康署: https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=1128&pid=1949 3. An Kao, Jui-Chung Hung*,Chih-Peng Huang. (2014). 支援向量迴歸建立血壓預測模型改善血壓量測的不準確性. The 31st Workshop on Combinatorial Mathematics and Computation Theory. 4. C Alexakis,M Rodrigues,R Saatchi,HO Nyongesa,C Davies,RH Ireland,ND Harris,SR Heller. (2006). A Knowledge-Based Electrocardiogram-Monitoring System for Detection of the Onset of Nocturnal Hypoglycaemia in Type 1 Diabetic Patients. Computer in Cardiology. 5. C. Alexakis,H. O. Nyongesa,R. Saatchi,N. D. Harris,C. Davies,C. Emery,R. H. Ireland,S. R. Heller. (2003). Feature extraction and classification of electrocardiogram (ECG) signals related to hypoglycaemia. Computers in Cardiology. 6. D. X. Guo, Y. Z. Shang, R. Peng, S. S. Yong, X. A. Wang*. (2015). Noninvasive Blood Glucose Measurement Based on NIR Spectrums and Double ANN Analysis. Journal of Biosciences and Medicines. 7. Guido Freckmann, M.D.,1 Sven Hagenlocher,1 Annette Baumstark, Ph.D.,1 Nina Jendrike, M.D.,1Ralph C. Gillen, Ph.D.,1Katja Rössner, M.Sc.,2 and Cornelia Haug, M.D.1. (2007). Continuous Glucose Profiles in Healthy Subjects under Everyday Life Conditions and after Different Meals. Journal of Diabetes Science and Technology. Diabetes Technology Society. 8. Ian H. Witten, Eibe Frank. (2005). Data Mining Practical Machine Learning Tools and Techniques. San Francisco: ELSEVIER. 9. Kamat, D. K. ,Bagul, Dhanashri ,Patil, P. M. (2014). Blood Glucose Measurement Using Bioimpedance Technique. Advances in Electronics. 10. L.Taneva, V.Hristov and F.Ibrahim. (2013). Noninvasive method for measuring blood glucose using MSP430x microcontroller. Int. J. Open Problems Compt. Math., Vol. 6, No. 2, June 2013. 11. LED Light Therapy. (無日期). 擷取自 https://balancespaboca.com/led-facial/ 12. Linh Lan Nguyen, Steven Su, Member, IEEE, and Hung T. Nguyen, Senior Member, IEEE. (2012). Identification of Hypoglycemia and Hyperglycemia in Type 1 Diabetic Patients Using ECG Parameters. 34th Annual International Conference of the IEEE EMBS. San Diego, California USA,. 13. Monte-MorenoEnric. (2011). Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means of machine learning techniques. Artificial Intelligence in Medicine 53 (2011), 頁 127– 138. 14. Nina Korlina Madzhi, Sarah Addyani Shamsuddin, Mohd Firdaus Abdullah. (2014). Comparative Investigation Using GaAs(950nm),GaAIAs (940nm) and InGaAsP (1450nm) Sensors for Development of Non-Invasive Optical Blood GlucoseMeasurement System. Proc. of the IEEE International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA). Kuala Lumpur, Malaysia. 15. R. Rox ANDERSON,B.S. AND JOHN A. PARRISH M.D. . (1981). The Optics of Human Skin. THE JOURNAL OF INV ESTIGATIV E DERMATOLOGY, 77:13-19, 198 1 . 16. RaschkaSebastian. (2016). Python 機器學習. (劉立民, & 吳建華, 譯者) 新北市: 博碩文化. 17. S. K. Dhar ,P. Biswas ,S. Chakraborty. (2013). DC impedance of human blood using EIS: An appraoch to non-invasive blood glucose measurement. Informatics, Electronics & Vision (ICIEV), 2013. 18. S. Ramasahayam,L. Arora,S. R. Chowdhury,M. Anumukonda. (2015). FPGA based system for blood glucose sensing using photoplethysmography and online motion artifact correction using adaline. 2015 9th International Conference on Sensing Technology (ICST). 19. Sang-Soo OakAroulPraveen. (2015). How to Design Peripheral Oxygen Saturation (SpO2) and Optical Heart Rate Monitoring (OHRM) Systems Using the AFE4403. Texas Instruments. 20. Su-Hong Chen,Cheng-Ting Lin,Tzung-Cheng Tsai,Yu-Hsiuan Chang. (2014年12月). The Energy Saving Application of Electricity. Journal of Taiwan Energy Volume 1, No. 5, December 2014, pp. 601-612. 21. T. Das、A. Ghosh、S. Guha、P. Basak. (2017). Early detection of diabetes based on Skin impedance Spectrogram and heart rate variability noninvasively. 2017 1st International Conference on Electronics, Materials Engineering and Nano-Technology (IEMENTech). IEEE. 22. This has the function for the Clarke Error Grid. (無日期). 擷取自 github: https://github.com/suetAndTie/ClarkeErrorGrid 23. TsaiYen-lung. (無日期). 擷取自 github: https://github.com/yenlung 24. Yamakoshi, K , Yamakoshi, Y. (2006). Pulse glucometry: A new approach for noninvasive blood glucose measurement using instantaneous differential near-infrared spectrophotometry. Journal of Biomedical Optics. 25. 工研院推出「即時自我監測血糖儀」技術 5年後上市. (無日期). 擷取自 https://tw.news.yahoo.com/%E5%B7%A5%E7%A0%94%E9%99%A2%E6%8E%A8%E5%87%BA-%E5%8D%B3%E6%99%82%E8%87%AA%E6%88%91%E7%9B%A3%E6%B8%AC%E8%A1%80%E7%B3%96%E5%84%80-%E6%8A%80%E8%A1%93-5%E5%B9%B4%E5%BE%8C%E4%B8%8A%E5%B8%82-103323269--finance.html 26. 机器学习实战之--regression. (無日期). 擷取自 https://blog.csdn.net/sunnyxiaohu/article/details/50967274 27. 江冠世. (2017). 非侵入式血糖量測技術專利分析. 國立台灣大學專利研究所學位論文. 28. 李仁貴、陳安雋、鄭文典. (2017). 穿戴式生醫量測原理與實驗-使用Android Studio開發. 新北市: 全華圖書. 29. 李凱平. (2017年November 月6日). 淺談聯合型用戶需量競價之用電預測技術. 擷取自 http://www.syscom.com.tw/ePaper_New_Content.aspx?id=634&EPID=241&TableName=sgEPArticle 30. 張家齊、蕭子健. (2014). 生醫訊號系統實作 LabVIEW&Biomedical System. 新竹市: 國立交通大學出版社. 31. 陳允傑. (2018). Python 資料科學與人工智慧應用實務. 台北市: 旗標科技. 32. 陳束弘,林政廷,蔡宗成,張語軒. (2014). 電力耗能負載預測與節能應用. Journal of Taiwan Energy Volume 1, No. 5, December 2014, 頁 601-612. 33. 陳冠堯. (2014). 以機器學習演算法對糖尿病及糖尿病前期的預測. 台北: 臺北醫學大學. 34. 游輝禎、徐志宏. (2013). 高齡化社會生活環境發展之研究. 內政部建築研究所. 35. 葉國賢、黃寶震 、陳怡良. (2010). 非侵入式血氧濃度量測. 2010 Conference on Information Technology and Applications in Outlying Islands . 36. 葉蕢誠. (2014). 應用支援向量迴歸於交通資料遺失值之差補:以固定式車輛偵測器資料為例. 台北: 淡江大學. 37. 機器學習實戰教程. (無日期). 擷取自 https://itw01.com/V3C8ET7.html
摘要: 嵌入式系統與開放式資源蓬勃發展,資訊科技與其數據伴與生活密不可分 ,尤其是科技照護更是許多健康照護者持續關注的議題。觀察醫療照護商品的發展與趨勢,國人普遍對APPLE Watch、心律手環(手錶)、健康照護APP相關產品具有較高的接受度與較佳觀感。而目前的遠距健康照護系統,則主要提供居家健康照護服務,藉由配戴傳感器將生理數據上傳至系統,藉由持續性的生理感測與資料庫應用,當訊號異常被檢出時系統立即發出警告,提醒患者與照護者即時介入病情管理。本研究模組化的整合各式感測器、單晶片與無線網路晶片,藉由感測器將生物特徵換能至電訊號再透過數位系統把訊號轉換成數據資料,而無線網路晶片的建置則提供數據傳輸服務,使系統具有便捷、高精度與高整合度的應用。 本研究藉由醫學電子與生物資訊科學的技術整合,經由實作與實驗嘗試收集心率、體溫與血糖值等數據,利用機器學習套件來訓練數據並導出模型,再對血糖濃度進行數據預測。透過不同迴歸分析方法,嘗試找出最佳解與最小誤差估測,以完成非侵入式血糖儀的建置目標。雖然無法採集實際臨床數據,但實驗仍採用具臨床意義的黃金標準「克拉克誤差網格」分析,最終實驗結果顯示,最佳迴歸與預測量化模型是SVR支援向量迴歸模型。對於長期監測血糖的用戶,此非侵入式血糖推估實作將提供友善的使用者體驗。
The embedded systems and open resources are booming, computer sciences and its data are inseparable. In particular, technological health care is a topic that many health care providers continue to pay attention to. Observing the development and trends of health care products, people generally have high acceptance and perception of APPLE Watch, Smart Bracelet Watch, and health care APP related products. Currently, remote health care system mainly provides home health care services, by uploading physiological data to the system, continuous physiological sensing and database application. When the signal is abnormally detected, the system reminds patients and caregivers to check the patient's condition immediately. This paper integrates and modularizes various sensors, MCU and wireless chip, and using the biosensors and bioelectronics to convert the electrical signal into data by digital systems. The establishment of wireless chip provides data transmission services, making an application are convenience, highly precise and integrated function. This experiment attempted to collect data such as heart rate, body temperature and blood glucose value, and used the machine learning kit to train the data and derive the model to predict the blood glucose. Through different regression analysis methods, tried to find the best solution and the minimum error of estimate to complete the goal of building a non-invasive blood glucose meter. Due to the limitation, the experiment uses the clinical significant gold standard 'Clark Error Grid' to analyze. The results show that the best regression and predictive quantitative models is SVR. For users who monitor blood glucose for a long time, this non-invasive blood glucose estimation implementation will provide a friendly user experience.
URI: http://hdl.handle.net/11455/98246
文章公開時間: 2021-11-02
Appears in Collections:資訊科學與工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館

Show full item record
 
TAIR Related Article
 
Citations:


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.