An euclidean distance based KNN computational method for assessing degree of liver damage

Liver is one of the vital organs of human body. It performs number of metabolic functions that are essential for living a healthy life. Early diagnosis of liver disease is a difficult task because the symptoms are more visible in later stages of the damage. Appropriate evaluation of patients becomes a challenge for clinicians which eventually make the disease more alarming. This study according aims to construct an intelligent computing method for classifying various degree of liver damage. Correct identification of degree of liver damage will help the physicians to give appropriate amount of dose to liver patients. For implementation, linear discriminant analysis (LDA), diagonal linear discriminant analysis (DLDA), quadratic discriminant analysis (QDA), diagonal quadratic discriminant analysis (DQDA), classification and regression tree (CART) and k-nearest neighbors (KNN) are deployed. Attained simulation results demonstrated that euclidean distance metric based KNN computational method outperforms all and has given remarkably good results with a prediction accuracy of 92.53%.