臺灣博碩士論文加值系統

現今呼吸器已經被廣泛的使用於重大手術及重症患者，藉由呼吸器的輔助使病患渡過呼吸衰竭所產生的致命危機，提供較多的機會接受緊急處置以挽救生命。呼吸器適用病患的需求與自主呼吸的能力主要分為侵入型及非侵入型兩種不同的型式，侵入型呼吸器會增加肺炎之發生率7％~41％與死亡率35％~90％，進而延長呼吸器使用天數，導致呼吸器脫離困難。非侵入型是不需氣管插管或氣切的無創傷性機械通氣，可降低肺炎發生率及其它併發症，降低慢性阻塞性肺疾病急性惡化患者插管和死亡率，其他如氣喘、拔管後、手術後、外傷後、肺水腫和等待肺臟移植的病人等，都是可能的受惠者。但在病人心跳血壓不穩定、意識混亂、氣管內痰過多等症狀時，若未能及時使用侵入型呼吸器，反而使病情拖延，甚至病情惡化。所以在侵入型呼吸器及非侵入型呼吸器的選擇評估十分重要。

本研究為輔助評估病人使用非侵入呼吸器的最佳時機，運用人工智慧技術，收集個案醫院病人的生理數據，透過倒傳遞類神經網路、粒子群最佳化演算法及C5.0決策樹，建構最佳預測模型。研究結果顯示，倒傳遞類神經網路結合C5.0決策樹之準確率為93.796%，醫學評估指標面積0.964，皆優於其他模型。倒傳遞類神經網路結合C5.0決策樹建構之規則，經醫師確認在臨床診斷上具有效性且符合文獻，在臨床上對非侵入呼吸器使用時機具有重要的參考價值。

Ventilations have been widely used in major surgeries and on critical illness patients at present time. They help patients through the fatal crisis due to respiratory failure, providing more opportunities for the sick and injured to receive emergency care so more lives can be saved. Ventilations, depending on patients’ needs and their ability to breathe on their own, include two different types—invasive and noninvasive. Invasive ventilations will increase probability of pneumonia from 7% to 41% and mortality from 35% to 90%; they would prolong the days needing ventilations thus leading to difficulty of weaning the dependence. Noninvasive ventilation delivers the mechanically-assisted breaths without the need for intubation, such as endotracheal tube or tracheotomy. It can lower the chance of infections like pneumonia and other complications and can reduce the incubation and mortality rate on patients with acute deterioration of chronic obstructive pulmonary diseases (COPD). For patients with asthma, after extubation, post-surgery, with pulmonary edema, or waiting for lung transplant, noninvasive ventilations can be very helpful. However, for those with symptoms like unstable heart beat and blood pressure, confusion, excess mucus in trachea, etc. it is critical to use invasive ventilations on a timely matter. Without it, symptoms can be delayed and deteriorated. Therefore, the assessment on the selection of invasive or noninvasive ventilations is extremely important.

This study aimed to find the best timing to use noninvasive ventilations for patients, using artificial intelligence technology, collecting physiological data of patients in the case hospital, though BPN network, PSO algorithm, and C5.0 decision tree, the most optimal predictive model is constructed. The results show the accuracy of 93.796% in the use of the BPN network combing with C5.0 decision tree, medical evaluation index area at 0.964; both figures are better than the other models. The rule generated from the BPN network combining with C5.0 decision tree, upon confirmation of physicians, has proved most effective in clinical diagnosis and matches with literature results as well. It has important referential values clinically for the timing to use noninvasive ventilations.

中文摘要 ......................i
Abstract ......................ii
誌謝 ......................iii
圖目錄 ......................ix
第一章 緒論......................1
1.1 研究背景與動機......................1
1.2 研究目的......................2
1.3 研究範圍與限制...................... 3
1.4 研究流程 ......................4
第二章 文獻探討 ......................6
2.1 呼吸器......................6
2.1.1 使用呼吸器主要目的......................6
2.1.2 使用呼吸器的時機......................7
2.2 呼吸器之分類 ......................8
2.2.1 侵入型呼吸器 ......................8
2.2.2 氣管插管可能之生理病理反應及創傷傷害........9
2.2.3 侵入型呼吸器常見併發症......................10
2.3 非侵入型呼吸器 ......................11
2.3.1 病人選取 ......................12
2.3.2 非侵入型呼吸器適用時機......................13
2.4 類神經網路 ......................16
2.4.1 類神經網路的基本架構 ......................16
2.4.2 類神經網路模式 ......................17
2.4.3 倒傳遞類神經網路 ......................18
2.4.4 倒傳遞類神經網路重要參數......................19
2.4.5 國內外類神經網路之相關醫學研究 ...............21
2.5 決策樹 ......................22
2.5.1 決策樹分類模式 ......................23
2.5.2 決策樹國內外相關研究 ......................25
2.6 粒子群演算法 ......................26
2.6.1 粒子群演算法簡介 ......................27
2.6.2 粒子群演算法演算步驟 ......................27
2.6.3 粒子速度更新法則 ......................28
2.6.4 粒子群最佳化演算法之相關研究......................30
第三章 研究方法 ......................31
3.1 研究對象 ......................31
3.2 研究架構 ......................31
3.3資料蒐集與處理 ......................34
3.3.1 資料蒐集 ......................34
3.3.2 資料處理 ......................34
3.4 變數的選擇 ......................35
3.5 建立模型 ......................36
3.5.1 K疊交叉驗證法......................36
3.5.2 倒傳遞類神經網路模型建立................37
3.5.3 粒子群最佳化模型建立................40
3.5.4 C5.0決策樹模型建立 ................43
3.5.5 倒傳遞類神經網路結合C5.0決策樹模型建立 ..44
3.5.6 粒子群最佳化演算法結合C5.0決策樹模型建立...45
3.6 模型評估 ................46
3.6.1 混亂矩陣表 ................46
3.6.2 ROC曲線 ................47
第四章 資料分析與研究結果 ................48
4.1 資料敘述統計 ................48
4.2 倒傳遞類神經網路模型 ................50
4.2.1 倒傳遞類神經網路最佳參數組合設定................50
4.2.2 倒傳遞類神經網路模型建構 ................56
4.2.3 倒傳遞類神經網路結果分析 ................57
4.3粒子群最佳化演算法模型 ................58
4.3.1粒子群演算法最佳參數組合設定 ................58
4.3.2 粒子群演算法結果分析 ................59
4.4 C5.0決策樹模型................ 60
4.4.1 C5.0決策樹模型建構 ................60
4.4.2 C5.0決策樹模型結果分析 ................61
4.5 倒傳遞類神經網路結合C5.0決策樹模型 ............63
4.5.1 倒傳遞類神經網路結合C5.0決策樹模型建構.............63
4.5.2 倒傳遞類神經網路結合C5.0決策樹結果分析............65
4.6 粒子群演算法結合C5.0決策樹模型 ................67
4.6.1 粒子群演算法結合C5.0決策樹模型建構................67
4.6.2 粒子群演算法結合C5.0決策樹模型結果分析.............69
4.7 模型評估 ................71
4.7.1模型準確率與ROC比較 ................71
4.7.2 最佳模型分類規則表 ................75
第五章 結論與建議 ................79
5.1 結論 ................79
5.2 建議 ................81
參考文獻 ................82
Extended Abstract
簡歷

1.王人澍、張寶源、熊雅意 (2008)，應用決策樹理論於中醫辨證－以慢性咳嗽為例，中西整合醫學雜誌，10卷，2期，頁25-33。
2.王正信、董慧萍、蕭春蘭、沈慧怡、何瓊文 (2008)，非侵襲性正壓通氣在重症治療的角色，秀傳醫學雜誌，8卷，3期，頁145-152。
3.台大醫院侵入性管路照護標準-氣管插管，http://ppt.cc/EBgY，，中華民國，九十九年三月二十日。
4.加護病房病人救星-呼吸器自動化時代來臨，中時新聞網，黃曼瑩，http://ppt.cc/zzx9，中華民國，九十八年十二月二十六日。
5.何子銘、盧瑜芬、許家瑋、白健佑、白璐、周雨青、孫建安、Thomas Wetter、林金定、楊燦、朱基銘 (2006)，運用三種資料探勘方法預測子宮頸癌存活情形之比較，台灣家庭醫學雜誌，16 卷，3期，頁192-203。
6.李俊德 (1998)，氧合指數及其臨床意義，中華民國呼吸照護簡訊，9卷，1期，頁46-50。
7.李琳 (2003) ，重症醫學與護理-呼吸器的臨床應用，頁99-108，藝軒圖書出版社，台北。
8.李禎祥、劉建衛 (2005)，縮短呼吸器相關肺炎的抗生素療程，並探討病人預後與減緩院內抗藥性菌株產生，感染控制雜誌，15卷，1期，頁53-57。
9.周知 (2010)，續寫生命的機械通氣，浙江省麗水市中心醫院醫院發展宣傳處。
10.林明傑、董子毅 (2008)，危險評估中ROC曲線在預測2×2表上與敏感度及特異度之關係，亞洲家庭暴力與性侵害期刊，4卷，2期，頁64 -74。
11.林恆毅、蘇千玲、楊式興、陳群互、陳建文、吳清平 (2005)，非侵入性正壓呼吸器對拔管失敗之影響，輔仁醫學期刊，3期，頁33-41。
12.林碧華、李俊德、陳寶貝 (2002)，年齡對呼吸器患者預後的影響，中華民國呼吸治療期刊，1卷，1期，頁61-67。
13.林碧華、陳寶貝、李俊德 (2009)，住院之慢性阻塞性肺疾病患使用非侵性正壓呼吸器之成效，呼吸治療雜誌，頁13-15。
14.美婦「鐵肺」維生58年-停電要了她命，wow今日新聞國際篇，http://ppt.cc/6V5r，中華民國，九十九年一月十六日。
15.徐正會、古文琦、郭許達、吳健樑 (2004)，呼吸器之分類與分析，中國機械工程學會第二十一屆全國學術研討會論文集，台北，中華民國，十一月二十六日，二十七日。
16.徐正會、古文琦、郭許達、吳健樑 (2004)，呼吸器之分類與分析，中國機械工程學會第二十一屆全國學術研討會論文集，台北，中華民國，十一月二十六日，二十七日。
17.連德正 (1995)，非侵襲性呼吸器，內科學誌，3卷，6期，頁26-34。
18.連德正 (1999)，治療急性呼吸衰竭之新利器：非侵襲性正壓呼吸器，內科學誌，10卷，5期。
19.陳銘樹、王建智、王麗雁 (2008)，應用決策樹演算法以探究高科技員工潛在的糖尿病之危險因子，健康管理學刊，6卷，2期，頁135-146。
20.陳麗帆、周繡玲、周雨青、楊燦、簡戊鑑、白璐、孫建安、曾知浩、朱基銘 (2009)，使用比較資料探勘演算法預測非小細胞肺癌患者預後因子、存活情形及其效能，腫瘤護理雜誌，9卷，2期，頁33-47。
21.傅家啟、蔡志文、陳啟昌、王安祥、盧宗斌、徐賢鈞 (2005)，脈衝耦合類神經網路於磁振影像之腦組織分割，中華放射醫誌，30卷，2期，頁83-94。
22.黃美玲、陳幸宜、陳貴琳、李雅雯 (2006)，「類神經網路輔助醫療診斷分類模式之建構」，中華民國品質學會第42 屆年會暨第12 屆全國品質管理研討會，台北，中華民國，十一月一日，二日。
23.温浩、郭崇慧 (2010)，利用粒子群優化的人臉特徵提取識別算法，西安交通大學學報，44卷，4期，頁48-51。
24.葉金標、童春芳、張巍獻 (2009)，醫療院所之門診失約預測：類神經網路之應用，臺灣公共衛生雜誌，28卷，5期，頁361-373。
25.葉怡成(2006)，類神經網路模式應用與實做，儒林圖書公司，頁1-51，台北。
26.認識氣管插管，國立成功大學醫學院附設醫院內科加護病房，http://ppt.cc/BgrK，中華民國，九十九年三月十四日。
27.蔡蕙如、柯明中、張偉斌、劉德明 (2007)，應用類神經網路與分類迴歸樹於肝癌分類模式，北市醫學雜誌，4卷，8期，頁658-667。
28.Antonelli M., Conti G.., Rocco M., Bufi M., Alberto R. (1998). “A comparison of noninvasive positive-pressure ventilation and convention mechanical ventilation in patients with acute respiratory failure”, N Engl J Med, Vol. 339, No.7, pp. 429-435.
29.Bethesda. (1990), “Value of the ventilation/perfusion scan in acute pulmonary embolism. Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). The PIOPED Investigators”, Jama,Vol. 20, pp.263-2753.
30.Bellone A., Monari A., Cortellaro F., Vettorello M., Arlati S. (2004). “Myocardial infarction rate in acute pulmonary edema: noninvasive pressure support ventilation versus continuous positive airway pressure”, Crit Care Med, Vol. 32, pp.1860-1865.
31.Berry, M., Linoff G. (1999), Mastering Data Mining, the Art & Science of Cutomer Relation Management, John Wiley, New York, USA.
32.Brochard L., Mancebo J., Wysocki M., Lofaso F., Conti G., Rauss A. (1995). “Noninvasive ventilation for acute exacerbation of chroni obstructive pulmonary disease”, N Engl J Med , Vol. 333, No.13, pp.817-822.
33.Chae, Y.M., Kim, H.S., Tark, K.C., Park, H.J., Ho S.H. (2003) “Analysis of healthcare quality indicator using data mining and decision support system”, Expert Systems with Applications, Vol. 24, pp.167-172.
34.Clerc, M. (1999), “The swarm and the queen: towards a deterministic and adaptive particle swarm optimization”, IEEE Congress on Evolutionary Computation, Washington, DC, pp. 1951-1957.
35.Clerc, M. (2004). “Discrete particle swarm optimization, illustrated by the Traveling Salesman Problem”, New Optimization Techniques in Engineering, Heidelberg,Germany: Springer, pp. 219-239.
36.Confalonieri M., Garuti G., Cattaruzza M.S., Osborn J.F., Antoneli M. (2005). “A chart of failure risk for noninvasive ventilation in patients with COPD exacerbation”, Eur Respir J, Vol. 25, pp.348-355.
37.Delen, D., Walker, G., Kadam, A. (2005) “Predicting breast cancer survivability: a comparison of three data mining methods”, Artificial Intelligence in Medicine, Vol. 34, pp.113-127.
38.Eberhart, R.C. and Shi, Y.(2001) “Particle Swarm Optimization: Developments, Applications and Resources”, IEEE Congress on Evolutionary Computation, Vol.1, pp. 81-86.
39.Elliott M. W. (2002). “Non-invasive ventilation in acute exacerbations of chronic obstructive pulmonary disease: a new gold standard?”, Intensive Care Med, Vol 28, pp. 1691-1694.
40.Freeman, J. A., Skapura, D. M. (1992). Neural Networks Algorithms, Applications, and Programming Techniques, Addison-Wesley, Reading, Michigan, USA.
41.Hastie, T., Friendman J., Tibshirani R. (2001). The element of statistical learning, Springer-Verlag, New Work, USA.
42.Heyland D.K., Cook D.J, Griffith L. (1999). “The attributable morbidity and mortality of ventilator-associated pneumonia in the critically ill patient. The Canadian critical trials group”, Am JRespir Crit Care Med, Vol. 159, pp.1249-1256.
43.Hill, N.S. (1986). “Clinical application of body ventilators”, Chest, Vol. 90, pp. 897-905.
44.Hillberg R.E., Johnson DC. (1997). “Noninvasive ventilation”, N Engl med, Vol. 337 , pp. 1746-1752.
45.Jasmer R.M., Luce J.M., Matthay M.A. (1997). “Noninvasive positive pressure ventilation for acute respiratory failure: underutilized or overrated?”, Chest, Vol. 111, pp.1672-1678.
46.John V. Peter., John L. Moran., Jennie Phillips-Hughes., David Warn. (2002). “Noninvasive ventilation in acute respiratory failure- A meta-analysis update”, Crit Care, Vol. 30, pp.555-562.
47.Keenan S.P., Kernerman P.D., Cook D.J., Martin C.M., McCormack D., William J.S, Sibbald W.J. (1997). “Effect of noninvasive positive pressure ventilation on mortality in patients admitted with acute respiratory failure: a meta-analysis”, Crit Care Med, Vol. 25, pp.1685-1692.
48.Kennedy, J., & Ebernart, R. (1995) “Particle swarm optization”, IEEE International Conference on Neural Network, Vol. 4, pp.1942-1948.
49.Kennedy, J., & Spears, W.M., (1998) “Matching algorithms to problems: an experimental test of the particle swarm and some genetic algorithms on the multimodal problem generator”, IEEE International Conference on Evolutionary Computation, pp. 39-43.
50.Kohavi R. (1995). “A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection”, International Joint Conference on Artificial Intelligence, pp. 1137-1145.
51.Markku Siermala., Martti Juhola. (2006). “Techniques for biased data distributions and variable classification with neural network applied to otoneurological datas”, Computer Methods and Programs in Biomedicine, Vol. 8, pp.28-36.
52.Meduri GU. (1996). “Noninvasive positive-pressure ventilation in patients with acute respiratory failure”, Clin in Chest Med, Vol. 17 , pp.513-53.
53.Merwe D. W. & Engelbrecht. (2003) “Data Clustering using Particle Swarm Optimization”,IEEE Conf Syst, Vol.1 ,pp.215-220.
54.Mosby. (2007). Benumof’s Airway Management, Principles and Practice.
55.Nava S., Rubini F., Zanotti E. (1994). “Survival and prediction of successful ventilator weaning in COPD patients requiring mechanical ventilation For more than21days”, EurRespir J, Vol. 7, pp.1645-1652.
56.Nava S., Ambrosino N., Clini E. (1998). “Noninvasive mechanical ventilation in the weaning of patients with respiratory failure due to chronic obstructive pulmonary disease”, Am Intern Med, Vol. 128, pp.721-728.
57.Perreault W.D., H.C. Barksdale. (1980) “A model-free approach for analysis of complex contingency data in survey research”, Journal of Marketing Research XVII November, pp.503–515.
58.Plant PK, Owen JL, Elliott MW. (2000). “Early use of noninvasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on generaI respiratory wards-a multicenter randomised controlled trial”, Lancet, Vol. 355, pp.1931-1935.
59.Quinlan J,R., (2003), “C5.0 online tutorial”, http://www.rulequest.com.
60.Rello J., Ollendorf D.A., Oster G. (2002). “Epidemiology and outcomes of ventilator-associatet pneumonia in a large US database”. Chest, Vol. 122, pp.2115-2121.
61.Rumelhart, D., Hinton, G.E., Williams, R.J. (1986) “Learning Internal Representations by Error Propagation”, MIT press.
62.Samar F., Ziad W.G., Edward R.H., Robert S.B., James K.S., (2006). “Use of noninvasive positive-pressure ventilation on the regular hospital ward: experience and correlates of success”, Respir Care, Vol. 51,No.11, pp.1237-1243.
63.Shi, Y. (2004). “Particle Swarm Optimization”, IEEE Connections, vol. 2, pp.8-13.
64.Stone, M. (1974). “Cross-Validatory choice and assessment of statistical predictions” , Journal of the Royal Statistical Society B, Vol. 36, pp. 111-147.
65.Szpurek D., Moszynski R., Smolen A., Sajdak S. (2005). “Artificial neural network computer prediction of ovarian malignancy in women with adnexal masses”, International Journal of Gynecology and Obstetrics, Vol. 89, pp.108-113.
66.Tasnim S, Deborah J Cook, Jill R, Christopher J Allen.(2003). Chest, Vol. 123, pp.2062-2073.
67.Teasdale G., & Jennett B. (1974). “Assessment of coma and impaired consciousness- A practical scale”, Lancet, Vol. 2, pp.81-84.
68.Thomas J Martin., Jeffrey D Hovis., Joseph P Costantino., Morris I Bierman., Michael P Donahoe. (2000), “A Randomized, Prospective Evaluation of Noninvasive Ventilation for Acute Respiratory Failure”, Am J Respir Crit Care Med, Vol. 161, pp.807-813.
69.Thomas M T Cheung., Loletta K-Y So., Arthur C W Lau., Loretta Y C Yam., Edwin Poon, Raymond W H Yung., K Y Yuen. (2004) “Effectiveness of Noninvasive Positive Pressure Ventilation in the Treatment of Acute Respiratory Failure in Severe Acute Respiratory Syndrome”,Chest, Vol. 126. No. 3, pp.670-674.
70.Ture, M., Kurt, I.., Kurum, A.T., Ozdamar, K. (2005) “Comparing classification techniques for predicting essentiall hypertension”, Expert System with Applications, Vol. 29, pp.583-588.
71.Venugopal,V., Baets, W. (1994). “Neural Networks and Statistical Techniques in Marketing Research: A Concept Comparison”, Marketing Intelligence & Planning, Vol.12, No. 7, pp.30-38.
72.Worth, A.P., Cronin, M.T.D. (2003) “The use of discriminant analysis, logistic regression and classification tree analysis in the development of classification models for human health effects”, Journal of Molecular Structure, Vol. 622, pp.97-111.
73.Xia,W.P., Wu Z. (2005). “An effective hybrid optimization approach for multi-objective flexible job-shop scheduling problems”, Computer and Industrial Engineering, Vol. 48, pp.409-4