Analysis and Comparison of Kidney Stone Detection using Parallel Piped Classifier and Bayesian Classifier with Improved Classification Accuracy

Abstract

Aim: The goal of this research is to use parallel piped classifiers and bayesian classifiers to predict and detect kidney stones. Materials and Methods: This investigation made use of a collection of data from Kaggle website. Samples were considered as (N=10) for parallel piped classifiers and (N=10) for bayesian classifiers according to clinicalc.com, total sample size calculated. The accuracy was calculated by using MATLAB with a standard data set. Pretest G power taken as 85 in sample size calculation can be done through clinical.com. Results: The accuracy (%) of both classification techniques are compared using SPSS software by independent sample t-tests. There is a significant difference between the two classification techniques. Comparison results show that innovative parallel piped classifiers give better classification with an accuracy of (83.5410%) than bayesian classifiers (71.1314%).There is a statistical significant difference between parallelepiped classifiers and bayesian classifiers. The parallel piped classifiers with p=0.007, p<0.05 significant accuracy (83.54%) showed better results in comparison to bayesian classifiers. Conclusion: The parallel piped classifiers appear to give better classification accuracy than the bayesian classifiers.