應用資料探勘技術於慢性腎臟病之預測   免費試閱

Application of Data Mining Techniques for Detection of Chronic Kidney Disease

慢性腎臟疾病是目前全世界主要的公共健康問題之一。根據臺灣腎臟醫學會資料顯示，臺灣洗腎發生率高居世界第1位，平均每1.2小時，就新增1名洗腎患者，推估慢性腎臟病第3～5期的病人，更高達115萬人，超越糖尿病的百萬人口，儼然成為新國民病（邱鼎鈺，2008）。如果能即早發現達到早期治療之效果，將可以改善末期腎臟病變不斷升高的情形，進而改善末期腎病變的早發性更可達到減少健保資源之耗費。本研究以南部某區域醫院之慢性腎臟個案管理病患資料為例，期望藉由資料探勘技術研究慢性腎臟病各分期演變及針對各資料數據做統計分析，找出變數之影響性及各統計數據之意義。針對本研究的發現提出相關之建議，提供臨床醫護人員協助慢性腎臟病患者各階段適切之醫療診治、護理衛教等服務。

Chronic renal disease is one of the most important public health problems all over the world. According to the Taiwan Society of Nephrology, the incident of dialysis in Taiwan ranked first in the world’s new patient of dialysis was added every 1.2 hours; the number of chronic renal disease patients from stage III to stage V is estimated to reach up to 1.15 million, exceeding the number of million of diabetes patients. End stage renal disease is usually slowly transformed from the chronic renal disease stage I to stage V. If we discover the symptom and put it into remedy early, we can stop the end stage renal disease from rising, and then improve the early onset of end stage renal disease so as to reduce the waste of health care recourses. This study refers to the cases of chronic renal disease patients managed by a southern regional hospital, expecting by data mining techniques for all stages of chronic kidney disease and the evolution of information and data for the statistical analysis to identify variables affecting the meaning of various statistical data. Against the findings of this study are some suggestions to provide clinical staff with appropriate medical treatment, care and health education services to chronic renal disease patients of all stages.