Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.
KMID : 1155220120370020099
Journal of the Korean Society of Health Information and Health Statistics
2012 Volume.37 No. 2 p.99 ~ p.104
Decision Tree Models for the Risk Factors on the Adjusted Upper Limit of Normal Serum Alanine and Asparate Aminotransferase Levels in Korea
Kwak Min-Jung

Abstract
Objectives: The upper limit of normal levels for ALT and AST are updated using recently suggested criteria. We analyze the effect of risk factors(age, BMI, total cholesterol, protein, albumin) on ALT and AST using decision tree models. Methods: The decision tree models are adopted to find most affecting factors for the upper limit of normal levels for ALT and AST. Also, the threshold points for each affecting factor are determined by decision tree models. Results: The decision tree models showed that the cases with BMI over 25 and total cholesterol over 180.5 are high risk group for ALT. The most influencing factors for AST are also total cholesterol and BMI with threshold points 208.5 and 25, respectively. Conclusions: The most affecting factors on the upper limits of normal levels of ALT and AST are BMI and total cholesterol as metabolic parameters
KEYWORD
Upper Limit, ALT, AST, Decision tree
FullTexts / Linksout information
Listed journal information
ÇмúÁøÈïÀç´Ü(KCI)