作者: Mashhour Bani Amer
关键词:
摘要: This paper presents an assessment of liver function using novel neuro-fuzzy model blood albumin level BA. The developed that is used to predict the BA consists four inputs: Asparate Aminotransferace AST, Alkaline Phosphate ALP, Total Bilirubin T. Bil. and Protein Prot., which are measured in any routine test. proposed was trained 211 data a root-mean square error RMSE 0.29 for 100 epochs achieved. performance validated 57 testing sets 0.34 correlation coefficient CC between predicted values statistically significant CC=0.83, ensures efficiency accuracy fuzzy predicting main clinical benefit this it improves capabilities diseases can be as integral part medical expert system denoted diagnosis disorders.