脈診是傳統中醫裡很重要的診斷疾病的方法之一，過去的脈診儀都是先由單點感測方式進行開發，但因為只有一個單點感測因此容易漏掉非常多的資訊。後來開始有了多點式的感測而增加許多資訊，但多點感測方式的脈診儀單價高且不便攜帶。因此本論文想利用單點感測方式，透過不同面向分析方法增加資訊量，來提高脈象判斷的正確性。已知市面上有一項產品是透過量測皮膚電阻(Galvanic Skin Response, GSR)的方法來判斷疾病，且脈診與皮膚電阻感測都可進行判斷疾病之依據。因此本論文根據12個GSR特徵、王唯工教授所提出來共振理論的光體積變化描記圖(Photoplethysmography, PPG)訊號頻率域C1-C10中的7個特徵(P1)以及PPG訊號時間域的8個特徵(P2)，藉由支援向量機(Support Vector Machine, SVM)演算法將參與者身上獲取到的27個特徵進行非弦脈與弦脈之辨別。在本文的實驗驗證方法中，我們將三種不同類型的特徵(GSR、P1及P2)進行7種排列組合，從實驗結果可知，若個別單獨使用GSR特徵、P1特徵及P2特徵，其結果分別僅有62.5%、62.5%及65%的辨別準確率；若將同時使用三種特徵其辨別準確率可達90%。從整體結果顯示，在本論文中不僅提出一個藉由結合GSR以及PPG特徵資訊之非弦脈與弦脈的辨別系統，同時也證實了此方法的結果皆比市面上單獨使用GSR、P1或P2等特徵都要來得好。故在未來的應用中，它將更適合被用於中醫脈象上之相關研究。

Pulse diagnosis is one of the most important methods for diagnosis in Traditional Chinese Medicine. In the past, pulse diagnosis instruments were single-point sensing which missed a lot of information. This issue has been resolved by latter developed multi-point sensing instruments accompanied by much higher cost and lacking of mobility. Therefore, we want to apply the single-point sensing method to increase the information volume through different aspect-oriented analysis methods to improve the accuracy of pulse judgment. There is a product diagnoses through measuring skin resistance (Galvanic Skin Response, GSR). Both pulse and skin resistance sensing can diagnose diseases. Therefore, according to the features of 12 GSR, 7 photoplethysmography (PPG) signal frequency domain in the C1-C10 (P1) and 8 PPG signal time domain (P2), the support vector machine (SVM) algorithm is applied to divide the 27 features, which acquired from the participants, into non-wiry and wiry pulse. In the experimental verification method of our work, we have arranged three different types of features (GSR, P1 and P2). According to the results, the GSR, P1 and P2 discrimination accuracy are 62.5%, 62.5% and 65% respectively. In the other hand, the discrimination accuracy can be up to 90% if three features are applied simultaneously. From the overall results, we proposed a non-wiry and the wiry pulse discriminating system combining GSR and PPG feature information and confirmed it is better than single GSR, P1 or P2 system. Therefore, in future applications, it will be more suitable for pulse research on Chinese medicine.

[1] Yu-Feng Chung, Yu-Wen Chu, Cheng-Ying Chung, Chung-Shing Hu, Ching-Hsing Luo “New Vision of the Pulse Conditions Using Bi-Sensing Pulse Diagnosis Instrument ”, IEEE, 2013
[2] 中谷義雄，（賴逢甲譯）良導絡理論的研究，日新文化出版社, 台灣, 1973
[3] Chun-Chang Lu, Wei-Hao Cheng, I-Jui Liu “Study for the Effect of Applied Diagnosis Temperature to Ryodoraku Acupuncture Readings”
[4] WANG Lei, SHANG Qian-qian, WANG Yi-qin, QIAN Peng, GUO Rui, YAN Hai-xia, ” Discussion on International Standard of English Translation for TCM Pulse Condition Name Terms ” Chinese Journal of Information on TCM, Mar.2017 Vol.24 No.3
[5] 許智傑，「新型中醫脈診系統之研發及應用」，中原大學生物醫學工程學
系，博士論文，2007年
[6] L.Y. Wei, and T. Winchester, “Electronic diagnose of arterial pulse,” J. of Med. Eng. & Technology, Vol. 9, No.4, pp. 183-186, 1985.
[7] 賴舜彬，「脈波測量暨分析系統之設計」，中原大學醫學工程研究所，碩
士論文，1989年。
[8] 張修誠：以脈搏諧波頻譜分析探討中醫臟象學說與器官共振理論之相 關性研究。台灣中國醫藥學院中國醫藥研究所博士論文，黃維三、王 唯工指導，1993
[9] Yu-Feng Chung, Yu-Wen Chu, Cheng-Ying Chung, Chung-Shing Hu, Ching-Hsing Luo “New Vision of the Pulse Conditions Using Bi-Sensing Pulse Diagnosis Instrument ”, IEEE, 2013
[10] 金母健康科技,金姆感測,http://www.jinmu.com.tw/sales/
[11] Christopher G. Scully, Jinseok Lee, Joseph Meyer, Alexander M. Gorbach, Domhnull Granquist-Fraser, Yitzhak Mendelson. “Physiological Parameter Monitoring from Optical Recordings With a Mobile Phone” IEEE Transactions on Biomedical Engineering, VOL. 59, NO. 2, Feb. 2012
[12] Domenico Grimaldi, Yuriy Kurylyak, Francesco Lamonaca, Alfonso Nastro. “ Photoplethysmography Detection by Smartphone’s Videocamera “ IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 15-17 September 2011, Prague, Czech Republic
[13] Sasivimon Sukaphat, Sukapon Nanthachaiporn, Kritsadayu Upphaccha, Pinta Tantipatrakul, “ Heart Rate Measurement on Android Platform “, IEEE, 2016
[14] Yi Chang Su, Ke Feng Huang, Yu Hsin Chang, Tsai Chung Li, Wei San Huang and Jaung Geng Lin. The effect of fasting on the pulse spectrum. American Journal of Chinese Medicine, 2000；28：409-417
[15] 中谷義雄，（賴逢甲譯）良導絡理論的研究，日新文化出版社, 台灣, 1973
[16] 王文憲，人體生理學， 希爾國際股份有限公司，1995
[17] M.E.A.D，經絡儀，http://www.medpex.com.tw/medpexCN/meadcn.html
[18] ARDK，經絡檢測系統，http://ardk168.blogspot.com/p/ardk.html
[19] 經絡科技經絡儀，https://www.meri-tech.com/blank-10
[20] 張鈺鑫：以多訊息脈波分析評估出血性休克大鼠之預後。台灣中國醫 藥大學中國醫學研究所博士論文，蘇奕彰指導，2006年
[21] 王唯工：氣的樂章，大塊文化出版，臺北，2002 年初版
[22] Wang Lin YY, Chang SL, Wu YE, Hsu TL, Wang WK.：Resonance-The missing phenomenon in hemodynamics. Circ. Res. 1991；69：246-24918.WANG Wei-Kung, HSU Tse-Lin1, BAU Jian-Guo, WANG-LIN
[23] D. KORPAS, J. HÁLEK, L. DOLEŽAL "Parameters Describing the Pulse Wave", Physiol. Res. 58: 473-479, 2009
[24] Sibylle Fallet, Mathieu Lemay, Philippe Renevey, C´elestin Leupi, Etienne Pruvot, Jean-Marc Vesin "Can one detect atrial fibrillation using a wrist-type
photoplethysmographic device?",Medical & Biological Engineering & Computing, 2018
[25] Gerardo Tusman, Cecilia M. Acosta, Sven Pulletz, Stephan H. Bohm, Adriana Scandurra, Jorge Martinez Arca, Matias Madorno, Fernando Suarez Sipmann “ Photoplethysmographic characterization of vascular tone mediated changes in arterial pressure: an observational study “, Journal of Clinical Monitoring and Computing, 2018
[26] 心律大師，腕式生理監控，http://taiwanscientific.com.tw/
[27] Han-Wook Lee, Ju-Won Lee, Won-Geun Jung, and Gun-Ki Lee, “The Periodic Moving Average Filter for Removing Motion Artifacts from PPG Signals”, International Journal of Control, Automation and Systems, vol. 5, no. 6, pp. 701-706, December 2007
[28] Wikipedia contributors. Dynamic time warping. Available: https://en.wikipedia.org/wiki/Dynamic_time_warping
[29] T.Kam. Ho, “ Random Decision Forests “, IEEE, 1995