透過您的圖書館登入
IP:3.83.87.94
  • 學位論文

利用『資料探勘技術』探討急診高資源耗用者之特性

A Research of Data Mining Applied to Emergency Department For High Resource-Used Patients .

指導教授 : 蘇喜
共同指導教授 : 黃興進(Hsin-Ginn Hwang)

摘要


為了改善急診室的擁擠情形、增進醫護人員之工作效率、提供全方位之緊急醫療救護服務,唯有充分了解急診病患之個人特質與其資源耗用之關係,藉此醫護人員方可提供專業化、個人化之醫療服務給急診病患,並針對資源耗用較高的特殊個案擬定策略方針予以重點管理。 歸納本研究之目的如下: 1.分析急診醫療高利用者(72小時再回急診者)之病患特質 2.分析病患特質與其醫療費用之耗用情形 3.探討滯留急診超過24小時之病患特質 4.實證資料探勘工具於醫院管理領域之實用性 5.挖掘出潛在的病患特質,以強化醫院之顧客關係管理 在研究方法上,本研究擬參考過去一貫使用之統計分析方法,更引用目前於企業管理上炙手可熱的『資料探勘技術』,其中包含叢集分析、購物籃分析、決策樹等方法來挖掘個案醫院急診資源高耗用者(如72小時內再回診、滯留時間超過24小時、高醫療費用之病人),期望從挖掘結果來偵查個案醫院之病患潛在特質。 「管理、資訊、統計」不僅是現在與未來醫療產業發展的重要主軸,更是下一世紀醫學研究的發展重點,因此本研究除了融合「管理理論」及「統計分析」外,更使用了目前企業管理廣泛應用之「資料探勘」技術,企圖挖掘出少數潛在之急診資源高耗用者,以提供急診室醫療照護人員擬定護理計畫、醫院高階管理者擬定決策之參考依據。

並列摘要


In order to improve the crowds of emergency department , increase the efficiency of medical staff , and supply comprehensive urgently medical service to patients , the only way is to understand the relationship between characteristics and expenses of patients in emergency department. In the other words , hospital can supply professional and personal service to patients who had some special needs by mining knowledge form large data of hospital. The research focuses on the following three kinds of high resource-used patients:patients who re-visit emergency during 72 hours,stay at emergency over 24 hours,and spend high medical expenditure in emergency department. The main purpose of this research is to explore the characteristics of high resource-used patients in Emergence department. At the same time, we can also make something come true like that : (1) proving the contribution of data mining technology in hospital management, (2) finding out potential characteristics of patient to enhance the customer relationship management in hospital. The methodology of this research is a popular technology applied in business management for many years , including profitability analysis, customers attraction, market segmentation, asset risk management, and cross selling. And then, this research use cluster analysis, basket analysis and decision tree to mine modeling of hidden patterns in large volumes of patient data and to transform such data into information and even knowledge. The conclusions of the research obtained are listed as follows: (1) High resource-used patients of the research in this hospital are much less than normal patients;even if this kinds of patients stay more times , spend more medical expenditure and re-visit emergency department during 72 hours. (2) The research suggests hospital can supply specific services to patients who have different medical needs;therefore , hospital can decrease the resource wasted by some patients.

參考文獻


急診照護指標執行手冊(民90)。財團法人醫院評鑑醫療品質策進會。
紀志賢、曾詠淑、林錫璋、陳澤生、蔡良敏(1999)。以資料探勘模式實施緊急救護量預測。慈濟醫學雜誌,11(4),337-342。
謝瑞芳(2001)。資料挖掘在飛機引擎零件失效預測之研究。竟宜大學資訊管理學系碩士論文。
蔡麗伶(民90)高雄市急診醫療之疾病分佈與再診率之分析。高醫公衛所碩士論文。
林俊漢(民92)。由醫院評鑑資料庫探討急診醫學品質。高醫公衛所碩士論文。

被引用紀錄


許哲源(2008)。以本體論為基礎的資料採礦方法於運動行銷之研究-以Adidas公司為研究對象〔碩士論文,淡江大學〕。華藝線上圖書館。https://doi.org/10.6846%2fTKU.2008.00937
陳佳玲(2006)。閱聽人媒體來源、廣告效果與媒體組合策略關聯性探勘之研究〔碩士論文,淡江大學〕。華藝線上圖書館。https://doi.org/10.6846%2fTKU.2006.01057
徐弘孝(2005)。以資料探勘技術改善半導體製程錫鉛凸塊高度良率之研究〔碩士論文,中原大學〕。華藝線上圖書館。https://doi.org/10.6840%2fcycu200500688
鄒淑英(2006)。利用資料探勘技術探討超長天住院病人之相關屬性〔碩士論文,元智大學〕。華藝線上圖書館。https://doi.org/10.6838%2fYZU.2006.00199
林永翔(2010)。資料切割排序法在關聯規則搜尋之應用-以台電事故維修系統為例〔碩士論文,國立屏東科技大學〕。華藝線上圖書館。https://doi.org/10.6346%2fNPUST.2010.00066

延伸閱讀