燒傷意外突發性的發生，除可能造成患者生命威脅及容貌毀傷與官能障礙外，所產生之醫療、心理、經濟、家庭等問題，對病人、 醫師與社會亦會造成影響。當燒傷患者住院，在早期首要目的除提高存活率外，住院日之評估也是具有重要意義的。因為住院日評估之結果對患者及其家屬壓力與負擔、臨床醫師療程上之安排規劃與醫院醫療資源管控都有相當程度之影響。
本研究採用倒傳遞類神經網路(Backpropagation Neural Network)建立燒傷患者住院日預測模型。實驗過程中將住院早期(急性期)階段分為「住院初期」與「療程初期」，住院後期階段則分為「出院準備期」，運用十摺交叉驗證法(10-fold Cross Validation)進行類神經網路網路學習。藉由分析住院早期的預測結果，以了解不同住院階段所採用之不同資料變數，對預測準確率的影響。同時亦參考住院後期之預測結果，以進一步驗證住院早期預測結果之參考價值與可行性，希望能在患者住院早期就有較佳之預測結果，藉此達到具有時效性之住院日管理機制。
實驗結果顯示，利用類神經網路建立之燒傷住院日預測模型，在住院早期階段(住院及療程初期)合理預測筆數百分比分別為50.37%及57.22%，與住院後期階段(出院準備期)之合理預測百分比62.13%相較，早期預測結果應已具有臨床之參考價值。

A burn injury is a disastrous trauma and can have very wide ranging impacts, including individual, family, and social. Burns patients generally have a long period of hospital stay whose accurate prediction can not only facilitate allocations of scarce medical resources but also help clinicians to counsel patients and relatives at an early stage of care. Besides prediction accuracy, prediction timing of length of hospital stay (LOS) for burn patients is also critical. Early prediction has profound effects on more efficient and effective medical resource allocations and better patient care and management.
Hence, the objective of this study is to apply a backpropagation neural network (BPNN) for predicting length of hospital stay (LOS) for burn patients at early stages of care. Specifically, we defined two early-prediction timing, including admission and initial treatment stages. Prediction timing at the admission stage is to predict a burn patient’s LOS when the patient is admitted into the Burns Unit. Prediction at the initial treatment stage refers to the timing right after the first surgery for burn wound excision and skin graft is performed (typically within 72 hours of injury if the patient’s condition allows). Experimentally, we evaluated the prediction accuracy of these two stages, using that achieved at the post-treatment stage (referring to the timing when all surgeries for burn wound excision and skin graft are performed) as benchmarks. The evaluation results showed that prediction LOS at the admission and the initial treatment stages could attain an accuracy of 50.37% and 57.22%, respectively. Compared to the accuracy of 62.13% achieved by the post-treatment stage, the performance reached by the initial treatment stage would consider satisfactory.

表 目 錄.............................VI
圖 目 錄...........................VIII
第一章 緒 論.................................1
第一節 研究背景 ..............................1
第二節 研究動機與目的.........................2
第三節 論文架構 ..............................4
第二章 文獻探討 ..............................5
第一節 燒傷臨床關鍵評估因素 ..................5
第二節 燒傷住院日研究相關文獻................11
第三節 類神經網路............................16
第三章 住院日預測時機及類神經模式之建立......20
第一節 預測時機 .............................20
第二節 住院日預測變數 .....................21
第三節 類神經網路建立 .....................23
第四章 實驗評估與結果 .....................26
第一節 病歷資料蒐集 .....................26
第二節 實驗步驟流程 .....................30
第三節 類神經網路參數決定 .............31
第四節 訓練驗證網路與結果分析 .............35
第五章 結 論 .............................43
第一節 研究特點與貢獻 .....................43
第二節 研究限制與未來研究方向 .............44
參考文獻 .............................46

中文文獻
[王雅如01] 王雅如，「住院日控制－中國醫藥學院附設醫院之經驗」，醫院雜誌，第28卷，第2期，2001年，p.78。
[白璐99] 白璐，「燒燙傷流行病學調查報告/燒燙傷資訊系統分析報告」，中華民國兒童燙傷基金會燒燙傷流行病學調查，1999年9月。
[沙田 03] 沙田救護支隊網站-評估的燒傷程度，摘自http://hk.geocities.com/stbadmedical01/0105.htm，2003年。
[金毓鴻81] 金毓鴻，「中華現代外科學全書-整形外科學」， 台灣商務印書館，1981年。
[葉怡成01] 葉怡成，「應用類神經網路」，儒林圖書公司，2001年。
[熊正輝 00] 熊正輝，「以類神經網路為工具預估癌症末期病人之存活」，財團法人安寧照顧基金會研究成果，2000年。
[劉致和00] 劉致和，「資料採礦(Data Mining)－燙傷住院病人資料庫可能的研究方向」，中華民國兒童燙傷基金會-燙傷專業新知，2000年。
[衛生署02] 行政院衛生署，「民國91年醫療機構現況及醫療服務量統計摘要」，摘自http://www.doh.gov.tw/statistic/data/醫療服務量現況及服務結果摘要/91摘要表/附表15.xls。
[韓揆74] 韓揆，「綜合醫院長期住院之初步研究」，中美技術季刊，第19卷，第1期，1974年，pp.11-15。
[韓揆82] 韓揆，「病人住院日之研究」，醫院雜誌，第15卷，1982年，p.18。
[羅淑芬01] 羅淑芬、黃秀梨、姚開屏、劉雪娥，「復健期燒傷病患照顧者壓力感受、社會支援及其相關因素」，臺灣醫學，第5卷，第2期，2001年，pp.28-37。
[羅華強01] 羅華強，「類神經網路MATLAB的應用－類神經網路的介紹」，清蔚科技，2001年，pp.1-10。
英文文獻
[A99] American Burn Association, “Guidelines for the Operation of Burn Units,” American Burn Association Publications,1999.(available at: http://www.ameriburn.org/pub/guidelinesops.pdf)
[BGS99] Barret, J. P., Gomez P., Solano I., Gonzalez-Dorrego, M., and Crisol, F. J., “Epidemiology and Mortality of Adult Burns in Catalonia,” Burns, Vol. 25, 1999, pp.325-329.
[CS95] Clayton, M. C. and Solem, L. D., “No Ice, No Butter: Advice on Management of Burns for Primary Care Physicians,” Postgraduate medicine, Vol. 97, No.5, 1995, pp.151-155, 159-160, 165.
[ELO78] Edlich, R. F., Larkman, N., O’Hanlon, J. R., Berry, R., Hiebert, J. M., Rodeheaver, G. T., and Edgerton, M. T., “A Proposed Modification of the American Burn Association’s Injury Severity Grading System,” Journal of the American College of Emergency Physicians (JACEP), Vol. 7, 1978, p.226.
[EB02] Estahbanati, H. K. and Bouduhi, N., “Role of Artificial Neural Networks in Prediction of Survival of Burn Patients—A New Approach,” Burns, Vol. 28, 2002, pp.579-586.
[GMM99] Gomez-Cia, T., Mallen, J., Marquez, T., Portela, C., and Lopez, I., “Mortality According to Age and Burned Body Surface in the Virgen del Rocio University Hospital,” Burns, Vol. 25, 1999, pp.317-323.
[GFM00] Gornez-Cla, T.,Franco, A., Mallin, J. M., Gimeno, M. A., Ferndndez-Mota, A., M`rquez, T., Portela, C., and Lrpez, I., “Mortality of the Paediatric Burn Population Treated at the Virgen del Rocio University Hospital, Seville, Spain in the Period 1968-1999,” Annals of Burns and Fire Disasters, Vol. 13, No. 2, June 2000,pp.67-72
[G03] Grider, D. J., Principles of ICD-9-CM Coding, American Medical Association Press, (2nd Edition), 2003.
[HYB02] Ho, W. S., Ying, S. Y., and Burd, A., “Outcome Analysis of 286 Severely Burned Patients: Retrospective Study,” Hong Kong Medical Journal, Vol. 8, No. 4, 2002, pp.235-239.
[JAD93] Jollis, J. G., Ancukiewicz, M., DeLong, E. R., Pryor, D. B., Muhlbaier, L. H., Mark, D. B., “Discordance Of Databases Designed For Claims Payment Versus Clinical Information Systems: Implications For Outcomes Research,” Annals of internal medicine, Vol. 119, 1993, pp.844-850.
[KA01] Kretschmann E. and Apweiler R., “Automatic Rule Generation for Protein Annotation with the C4.5 Data-mining Algorithm Applied on Peptides in Ensembl,” German Conference on Bioinformatics, 2001.
[LAR00] Laria, A. R., Alaghehbandan, R., and Nikui, R., “Epidemiological Study of 3341 Burns Patients During Three Years in Tehran, Iran,” Burns, Vol. 26, 2000, pp.49-53.
[L95] Latarjet, J., “A Simple Guide to Burn Treatment,” Burns, Vol. 21, 1995, pp.221-225.
[MJW97] Mertens, D. M., Jenkins, M. E., and Warden, G. D., “Outpatient Burns Management,” Nursing Clinics of North America, Vol. 32, 1997, pp.343-364.
[RHW86] Rumelhart,D. E., Hinton, G. E., and Williams, R. J., “Learning Internal Representations by Back-propagating Errors,” Parallel Distributed Processing: Explorations in the Microstructure of Cognition, MIT Press, Vol. 1, 1986, pp.318-362.
[RST98] Ryan, C. M., Schoenfeld, D. A., Thorpe, W. P., Sheridan, R. L., Cassem, E. H., and Tompkins, R. G., “Objective Estimates of the Probability of Death from Burn Injuries,” New England Journal of Medicine, Vol. 338, 1998, pp.362-366.
[S98] Saffle, J. R., “Predicting Outcomes of Burns,” New England Journal of Medicine, Vol. 338, 1998, pp.387-388.
[SAG93] Sawhney, C. P., Ahuja, R. B., and Goel, A., “Burns in India-epidemiologyand Problems in Management,” Indian Journal of Burns, Vol. 1, 1993, pp.1-4.
[SDI93] Stavropoulou, V., Daskalakis, J., and Ioannovich, J., ” A New Prognostic Burn Index,” Annals of Burns and Fire Disasters, Vol. 6, No. 2, 1993.
[SLB96] Still, J. M. JR., Law, E. J., Belcher, K., and Thiruvaiyarv, D., “Decreasing Length of Hospital Stay by Early Excision and Grafting of Burns,” Southern Medical Journal, June 1996.
[THE82] Tobiasen, J., Hiebert, J., and Edlich, R., “A Practical Burn Severity Index,” Burn Care, Vol. 3, 1982, p.229.
[WS00] Walczak, S. and Schart, J. E., “Reducing Surgical Patient Costs through Use of An Article Neural Network to Prediction Transfusion Requirement Requirements,” Decision Support Systems, Vol. 30, No. 2, 2000, pp.125-138.
[WBS97] Wong, B. K., Bonovich, T. A., and Selvi, Y., “Neural Network Applications in Business: A Review and Analysis of the Literature (1988-95),” Decision Support Systems, Vol. 19, 1997, pp. 301-320.
[WN95] Wong, M. K. and Ngim, R. C. K., “Burns Mortality and Hospitalization Time—A Prospective Statistical Study of 352 Patients in an Asian National Burn Center,” Burns, Vol. 21, 1995, pp.39-46.
[XDS97] Xhu, M., Donelan, M., and Sheridan, R., “The Index of Deep Burn Injury— An Analysis of 66 Extremity Sites in 15 Children,” Burns, Vol. 23, 1997, pp.11-14.