臺灣博碩士論文加值系統

隨著台灣近年全民健康保險制度發放經費緊縮、人口結構老化、國人對醫療品質要求提高等多重因素之下，各個醫療機構的營運成本提高、虧損風險增加，醫療機構勢必得對支出和醫療資源管理更為謹慎，其中又以病床資源的排他性、照護人力配置最為醫療機構所重視，而病患的住院天數又與病床資源配置息息相關，因此本研究將以馬偕醫院心臟內科住院申報紀錄為研究資料，並著重於找出影響病患住院天數的顯著因素、以倒傳遞類神經網路預測住院天數的可行性、不同住院階段對預測結果的影響以及不同網路架構對預測結果的影響，並建立出一個能用病患入院資料就可預測住院天數區間之類神經網路模型，期望能透過類神經網路強大的適應性以及預測能力，幫助醫療院所提早掌握病患住院天數、對病床資源做更好的安排，讓醫療機構能將有限的資源做最大的發揮。

The major concern for most hospitalists in Taiwan is to serve the aging population with the limited healthcare budget provided by the National Health Insurance Administration. As hospital beds are scarce, high occupancy rates are often preferred. To develop efficient admission policy and optimize bed management, it would be beneficial to investigate the critical factors which might determine the length of stay (LOS) in the early stage of admission. This research is to use artificial neural network (ANN) models to predict the LOS for patients in cardiology department during the pre-admission stage. After comparing with the regression model, the ANN models are able to predict patients with longer LOS.

摘 要 i
ABSTRACT ii
誌 謝 iii
目 錄 iv
表目錄 vi
圖目錄 vii
第一章　導論 1
1.1 研究背景與動機 1
1.2 研究目的 7
1.3 論文架構 8
第二章　文獻探討 10
2.1 住院天數相關文獻 10
2.2 資料探勘 13
2.3 學習演算法 15
2.3.1 非監督式學習 15
2.3.2 監督式學習 16
2.3.3 決策樹 17
2.3.4 支持向量機 18
2.3.5 最近鄰居分類法 20
2.4 類神經網路 20
2.4.1 倒傳遞類神經網路 23
2.5 類神經網路預測相關文獻 29
第三章　研究方法 32
3.1 資料收集 32
3.2 資料前處理 33
3.3 變數選擇 35
3.4 網路架構 39
3.5 結果評估 41
第四章　結果實證和分析 44
4.1 實驗平台 44
4.2 實驗步驟 44
4.2.1 偏離值篩選 45
4.2.2 因子分析 49
4.2.3 預測模式 51
4.2.4 樣本切割 52
4.3 倒傳遞類神經網路參數與回歸方程式 54
4.3.1 倒傳遞類神經網路參數 54
4.3.2 線性回歸網路方程式 57
4.4 實驗結果 59
4.4.1 主診斷間的差異 62
4.4.2 住院階段的差異 63
4.4.3 線性回歸與類神經網路預測效果之差異 63
第五章　結論與未來建議 64
5.1 結論 64
5.2 後續研究及未來建議 65
參考文獻 67
附錄 72

1.D. W Aha, Lazy Learning, Washington D.C., USA: Springer Science+Business Media, 1997.
2.K. Bache and M. Lichman，「Iris Data Set」，UCI Machine Learning Repository，http://archive.ics.uci.edu/ml/datasets/Iris，1988，2014年12月造訪。
3.Chen, C.-L, &;quot;A Study of Patients with Higher Medical Expenses in DRG 124 - A Regional Hospital as Case,&;quot; 2014 International Symposium on Computer, Consumer and Control, Taichung, 2014, pp. 938-941.
4.S. D. Culler, D. S. Jevsevar, K. G. Shea, K. K.Wright, and A. W. Simon, &;quot;The Incremental Hospital Cost and Length-of-Stay Associated with Treating Adverse Events Among Medicare Beneficiaries Undergoing TKA,&;quot; The Journal of Arthroplasty, 2014.
5.J. E. Dayhoff and J. M. DeLeo, &;quot;Artificial Neural Networks Opening the Black Box,&;quot; Conference on Prognostic Factors and Staging in Cancer Management: Contributions of Artificial Neural Networks and Other Statistical Methods. Arlington, Virginia: American Cancer Society, 2001, pp. 1615-1635.
6.M. Fathi, M. Mohebbi, and S. M. Razavi, &;quot;Application of Image Analysis and Artificial Neural Network to Predict Mass Transfer Kinetics and Color Changes of Osmotically Dehydrated Kiwifruit,&;quot; Food Bioprocess Technol, no.4, 2011, pp. 1357-1366.
7.Z. Ghahramani, Unsupervised Learning. In Advanced Lectures on Machine Learning. London, UK: Springer-Verlag, 2004.
8.E. Grossi, A. Mancini and M. Buscema, &;quot;International experience on the use of artificial neural networks in gastroenterology,&;quot; Digestive and Liver Disease, vol.39, no.3, 2007, pp. 278-285.
9.HebbDonald, The Organization of Behavior, New York: Wiley &; Sons, 1949.
10.R. Hecht-Nielsen, &;quot;Theory of the Backpropagation Neural Network,&;quot; International Joint Conference on Neural Networks, Washington, DC, 1989, pp. 593-605.
11.IBM, SPSS Statistics 21 Help Documentation, USA: IBM Corporation, 2012.
12.D. P. Janssen, L. Noyez, C. Wouters and R. M. Brouwer, &;quot;Preoperative prediction of prolonged stay in the intensive care unit for coronary bypass surgery,&;quot; European Journal of Cardio-thoracic Surgery, vol.25, 2004, pp.203-207.
13.Q. Jarosz，「Neuron Hand-tuned」，Wikipedia，http://commons.wikimedia.org/wiki/File:Neuron_Hand-tuned.svg，2009，2014年11月造訪。
14.T. Kohonen, &;quot;The Self-Organizing Map,&;quot; Proceedings of the IEEE, vol.78, no.9, 1990, pp. 1464-1480.
15.S. B. Kotsiantis, &;quot;Supervised Machine Learning: A Review of Classification Techniques,&;quot; Informatica, vol.39, pp. 249-268.
16.T.-S. Lim, W.-Y. Loh, and Y.-S. Shih, &;quot; A Comparison of Prediction Accuracy, Complexity, and Training Time of Thirty-Three Old and New Classification Algorithms,&;quot; Machine Learning, 2000, pp.203-228.
17.C.-L. Lin, P.-H Lin, L.-W Chou, S.-J. Lan, N.-H. Meng, S.-F. Lo, and H.-D. I. Wu, &;quot;Model-based Prediction of Length of Stay for Rehabilitating Stroke Patients,&;quot; Journal of the Formosan Medical Association, vol.108, no.8, 2009, pp. 653-662.
18.MathWorks，「Multilayer Neural Network Architecture」，MathWorks Documentation，http://www.mathworks.com/help/nnet/ug/multilayer-neural-network-architecture.html，2015，2015年6月造訪。
19.A. McAfee, E. and Brynjolfsson, &;quot;Big Data: The Management,&;quot; Harvard Business Review, 2012, pp. 59-68.
20.W. S. Mcculloch, and W. Pitts, &;quot;A Logical Calculus of the Ideas Immanent in Nervous Activity,&;quot; Bulletin of Mathmatical, Biophysics, vol.5, no.4, 1943, pp. 115-133.
21.M. F. Meller, &;quot;A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning,&;quot; Neural Networks, no.6, 1993, pp. 525-533.
22.A. Mellit, and A. M. Pavan, &;quot;A 24-h forecast of solar irradiance using artificial neural network: Application for performance prediction of a grid-connected PV plant at Trieste, Italy,&;quot; Solar Energy, vol.84, 2014, pp. 807-821.
23.M. R. Mickey, O. J. Dunn, and A. V. Clark, &;quot;Note on the Use of Stepwise Regression in Detecting Outliers,&;quot; Computers and Biomedical Research, no.1, 1966, pp. 105-111.
24.B. A. Mobley, R. Leasure, and L. Davidson, &;quot;Artificial neural network predictions of lengths of stay,&;quot; HEART &; LUNG, 1995, pp. 251-256.
25.K. H. Nagarsheth, S. S. Gandhi, R. E. Heidel, S. J. Kurek, and C. Angel, &;quot;A Mathematical Model to Predict Length of Stay in Pediatric ATV Accident Victims,&;quot; Journal of Surgical Research, vol.171, 2011, pp. 28-30.
26.F. Rosenblatt, &;quot;The Perceptron: A Probalistic Model For Information Storage And Organization In The Brain,&;quot; Psychological Review, vol.65, no.6, 1958, pp. 386-408.
27.D. E. Rumelhart, G. E. Hinton, and R. J. Williams, &;quot;Learning Internal Representations by Error Propagation,&;quot; Parallel distributed processing: Explorations in the microstructure of cognition, vol. 1, 1986, pp. 319-361.
28.T. M. Schmelzer, A. E. Gamal Mostafa, S. M. Camp, K. W. Kercher, T. S. Kuwada, and B. T. Heniford, &;quot;Factors Affecting Length of Stay Following Colonic Resection,&;quot; Journal of Surgical Research, vol.146, 2008, pp. 195-201.
29.N. Spratt, Y. Wang, C. Levi, K. Ng, M. Evans, and J. Fisher, &;quot;A Prospective Study of Predictors of Prolonged Hospital Stay and Disability After Stroke,&;quot; Journal of Clinical Neuroscience, vol.10, no.6, 2003, pp. 665-669.
30.P.-N. Tan, M. Steinbach, and V. Kumar, Introduction to Data Mining. Boston, USA: Pearson Education, 2006.
31.P. J. Werbos, Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. PhD thesis, Harvard University, Cambridge, 1974.
32.D. J. Whellan, X. Zhao, A. F. Hernandez, L. Liang, E. D. Peterson, D. L. Bhatt, G. C. Fonarow, &;quot;Predictors of Hospital Length of Stay in Heart Failure: Findings from Get With the Guidelines,&;quot; Journal of Cardiac Failure, vol.17, no.8, 2011, pp. 649-656.
33.Yu-Hsin Liu, Y.-C. C., &;quot;Impact of the Diagnosis Related Groups Prospective Payment System on the Profitability of Hospitals in Taiwan,&;quot; Journal of Medicine and Health, vol.2, no.2, 2013, pp. 23-34.
34.內政部統計處，「統計報告─重要參考指標」，內政部統計處網站，http://www.moi.gov.tw/stat/index.aspx，2014，2014年12月造訪。
35.宋晧遠，應用類神經網路在入院階段預測心血管患者住院天數，碩士論文，國立台北科技大學工業工程與管理系，台北市，2012。
36.郭仕堯、蕭樂群、張原賓，「類神經網路於飛航網路運量預測之應用」，航空、太空及民航學刊系列，第四十二卷，第一期，2010，第67-72頁。
37.黃宏斌，高雄港轉口貨櫃運量預測之研究─以類神經網路為預測模式，碩士論文，國立海洋大學航運管理學系，基隆市，2001。
38.蔡惠喻、余銘忠，高雄港轉口貨櫃量之運量預測，碩士論文，國立高雄應用科技大學企業管理研究所，高雄市，2012。
39.衛生福利部中央健康保險署，「102年全民健康保險統計」，
衛生福利部中央健康保險署網站， http://www.nhi.gov.tw/webdata/webdata.aspx?menu=17&;menu_id=1023&;WD_ID=1043&;webdata_id=4639，2014，2014年10月造訪。
40.衛生福利部中央健康保險署，「主題專區─DRGs支付制度」，
衛生福利部中央健康保險署網站， http://www.nhi.gov.tw/webdata/webdata.aspx?menu=17&;menu_id=1027&;webdata_id=937，2014，2014年10月造訪。