利用人工嗅覺的技術協助醫生了解病患、輔助醫生判斷病患的生理狀況，是目前很有潛力的生醫電子應用。由於氣喘與慢性肺阻塞兩種疾病之病徵極為相似，若能夠早期偵測、區分，在臨床上將有應用的價值。本研究針對這個主題，採用了化學感測器陣列與患者呼出之氣體反應，取得了反應訊號後，首先藉由訊號前處理技術，消除基線飄移及樣本間的濃度差異。接下來，氣體的辨識流程分為以下幾個部分：一、先行辨識該氣體是否為病患所呼出；二、若辨識結果為病人氣體，則分析該名患者罹患何種疾病(氣喘或慢性肺阻塞)；最後，為該名病患所罹患的疾病分析其患病嚴重度。辨識方法則是利用主成分分析法結合線性判別分析降低資料維度，減少資料點運算量；再使用支持向量機或K個最近鄰居法兩者相互比較，再以混淆矩陣觀察之。除了以上辨識分析外，我們更進一步地觀察主成分分析法降維時的各感測器權重，並以感測器之目標氣體與氣喘及慢性肺阻塞兩種患者呼出的常見揮發性有機化合物比較之。本篇論文的辨識結果發現，經由以上流程之辨識結果均達八成以上，且當我們在辨識疾病以主成分分析法回推降維的感測器權重時，發現其中一個權重較高的感測器TGS2620，該感測器之目標氣體二甲苯，是分辨慢性肺阻塞患者與氣喘患者呼出的氣體中，明顯不同的揮發性化合物之一。

The use of artificial olfactory technology to assist medical diagnosis is a promising domain for electronic applications. Since bronchial asthma (BA) and chronic obstructive pulmonary disease (COPD) have very similar symptoms, being able to detect and discriminate them automatically would have positive impact in the clinics. In this research, we focused on this topic and used a chemical sensing array to detect the gas exhaled from the patients. Signal preprocessing techniques were applied to remove the baseline drift and normalize the influence of concentration between samples. Then, methods for gas classification involved the following steps: first, we determined whether the gas was produced by a patient who suffered from either kind of the diseases. Secondly, if the classification result was positive, we applied further analysis to tell what kind of disease the patient had. Finally, we analyzed the severity of the patients. The method of classification was based on principal component analysis (PCA) and linear discriminant analysis to reduce the dimension of data. After that, we compared the result of the support vector machine and the K nearest neighbor method to achieve the best performance. We used the receiver operating curve and the area under the curve to assess the reliability and validity of the classifier. In addition, we observed the weights of the sensors in reduced dimensions of PCA and checked whether the main components of the analyzed result agreed with the list of commonly observed volatile organic compounds in the gas exhaled by patients of COPD and BA. The results of the identification all have a recognition rate of above 80%. Also, PCA-based analyses indicate that xylene, which is one of the volatile organic compounds found in gas exhaled by COPD patients, is a key compound that enable discrimination of COPD and BA.

摘要 I
Abstract II
誌謝 III
第一章 緒論 1
1.1 生物嗅覺機制 1
1.2 電子鼻系統介紹 2
1.3 慢性肺阻塞(COPD)與氣喘(BA)簡介 2
1.4 研究動機與文獻探討 4
第二章 實驗裝置與資料擷取流程 7
2.1 實驗設計 7
2.2 臨床資料取得方式及使用資料介紹 9
2.3 實驗方法與流程 11
第三章 分析方法 13
3.1 訊號前處理 13
3.1.1基線操作 14
3.1.2正規化 15
3.2 特徵萃取演算法 15
3.2.1 主成分分析 (Principal components analysis, PCA) 16
3.2.2線性判別分析 (Linear Discrimination Analysis, LDA) 17
3.3 分類器 21
3.3.1 K個最近鄰居法(k-Nearest Neighbors, KNN) [25] 22
3.3.2 支持向量機(Support Vector Machine, SVM) [27] 23
3.4分類效果評估方法 26
3.4.1 交叉驗證(Cross-Validation) [28] 26
3.4.2 混淆矩陣(confusion matrix)[29] 27
第四章 結果分析與討論 28
4.1偵測疾病 29
4.1.1辨識率 30
4.1.2 偵測疾病之混淆矩陣 31
4.2辨識疾病 32
4.2.1辨識率 33
4.2.2 辨識疾病之混淆矩陣 34
4.3病情嚴重度分群 34
4.3.1 慢性肺阻塞(COPD)之病情嚴重度分析 35
4.3.2 氣喘(BA)之病情嚴重度(FEV1)分析 38
4.4病情嚴重度之線性回歸討論 40
4.5感測器權重分析 41
4.5.1 辨識疾病之權重分析 41
4.5.2 病情嚴重度之權重分析 43
第五章 結論與未來發展 46
參考文獻 47
附錄 50
A.1偵測疾病線性判別分析之感測器權重 50
A.2辨識疾病線性判別分析之感測器權重 51
A.3 慢性肺阻塞病情嚴重度線性判別分析之感測器權重 53
A.4 氣喘病情嚴重度線性判別分析之感測器權重 54

[1] 林怡文, “一「嗅」萬千的巧妙,”科學人雜誌, 2012.
[2] W. Wilkens and J. Hartman, “An electronic analog for the olfactory processesa,” Journal of Food Science, 1964.
[3] T. Pearce, S. Schiffman, H. Nagle, and J. Gardner, “Handbook of machine olfaction: electronic nose technology,” 2006.
[4]彰化基督教醫院,“慢性肺阻塞(COPD). ”[online]. http://www2.cch.org.tw/lungcancer/COPD.htm
[5] G. Konvalina and H. Haick, “Sensors for breath testing: from nanomaterials to comprehensive disease detection,” Acc. Chem. Res., 2013.
[6] P. Jain and R. Kushwaha, “Wireless gas sensor network for detection and monitoring of harmful gases in utility areas and industries,” Sixth Int. Conf. Sens. Technol, 2012.
[7] 韩东海，江琳琳，潘磊庆，屠康，杨虹贤，钱钊，, “基于电子鼻对水蜜桃货架期评价的研究,” 食品科学, 2010.
[8] C. Di Natale and A. Macagnano, “Lung cancer identification by the analysis of breath by means of an array of non-selective gas sensors,” Biosens and Bioelectronics, 2003.
[9] S. P. Dragonieri S, Annema JT, Schot R, van der Schee MP, SpanevelloA, Carratú P, Resta O, Rabe KF, “An electronic nose in thediscrimination of patients with non-small cell lung cancer and COPD,” Lung Cancer, 2009.
[10] “Figaro Sensor.”[Online]Available:http://www.figarosensor.com/.
[11] 楊廷然, 「利用多標籤分類器實現電子鼻混合氣體識別方法之研究」, 國立清華大學電機工程研究所, 碩士論文, 2014.
[12] S. Marco and A. Gutiérrez-Gálvez, “Signal and data processing for machine olfaction and chemical sensing: a review,” Sensors Journal, IEEE, 2012.
[13] 許柏安, 「快速混合氣體辨識方法之研究」, 國立清華大學資訊工程研究所, 碩士論文, 2012.
[14]王家銘, 「利用樣式識別實現電子鼻肺炎偵測」, 國立清華大學電機工程研究所, 碩士論文, 2013.
[15] A. Fort, N. Machetti, and S. Rocchi, “Tin oxide gas sensing: comparison among different measurement techniques for gas mixture classification,” Instrumentation and Measurement, 2003.
[16] F. Acevedo and S. Maldonado, “Probabilistic support vector machines for multi-class alcohol identification,” Sensors Actuators B B: Chemical, 2007.
[17] C. Chang and C. Lin, “LIBSVM: A library for support vector machines,” ACM Trans. Intell. Systems and Technology (TIST)., 2011.
[18] 優活健康網,“何謂肺功能檢查？　注意事項、禁忌報你知. ”[online]. http://www.uho.com.tw/sick.asp?aid=13225
[19] R. Gutierrez-Osuna, “Pattern analysis for machine olfaction: a review,” Sensors Journal, IEEE, 2002.
[20] R. Gutierrez-Osuna and H. Nagle, “A method for evaluating data-preprocessing techniques for odour classification with an array of gas sensors,” Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions, 1999.
[21] G. Hughes, “On the mean accuracy of statistical pattern recognizers,” Inf. Theory, IEEE Trans., 1968.
[22] S. Wold, K. Esbensen, and P. Geladi, “Principal component analysis,” Chemometrics and intelligent laboratory systems, 1987.
[23] J. Ye, “Least squares linear discriminant analysis,” Proc. 24th Int. Conf. Mach. Learn., 2007
[24] P. Belhumeur, “Eigenfaces vs. fisherfaces: Recognition using class specific linear projection,” Pattern Analysis and Machine Intelligence, IEEE Transactions, 1997.
[25] N. Altman, “An introduction to kernel and nearest-neighbor nonparametric regression,” Am. Stat., 1992.
[26] D. Stork, R. Duda, and P. Hart, “Pattern classification,” IO/Il7, 2001.
[27] C. Burges, “A tutorial on support vector machines for pattern recognition,” Data Min. Knowl. Discov., 1998.
[28] R. Kohavi, “A study of cross-validation and bootstrap for accuracy estimation and model selection,” Ijcai, 1995.
[29] T. Fawcett, “An introduction to ROC analysis,” Pattern Recognit. Lett., 2006.
[30] ROC曲线（受试者工作特征曲线）分析详解[online]. http://www.seekbio.com/biotech/soft/2011/v820333248.html