Authors: Indrajit Mandal N Sairam
Publish Date: 2012/02/12
Volume: 36, Issue: 5, Pages: 3353-3373
Abstract
This paper presents more accurate and reliable computational methods for aiding the treatment of people with coronary artery disease New techniques are introduced for improved evaluation and distinguish cardiac disease affected patients from the healthy controls Experiments are conducted with high level of error tolerance rate and confidence level at 95 and 99 and established the results with corrected Ttests based on comparison of various performance measures Normal kernel density estimator is used for visual distinction of cardiac controls A new ensemble learning method comprising of Bayesian network as classifier and Principal components method as the projection filter with ranker search is used for the relevant feature selection Analysis of each model is performed and discusses major findings and concludes with promising results compared to the related works Multiple Correspondence analysis is used for exploring heart disease variable’s relationships Robust machine learning algorithms used are Rotation forests MultiBoosting Sparse multinomial logistic regression for better performance with fine tuning of their involved parameters The work aims at improving the software reliability and quality of diagnosis of cardiac disease with robust inference system To the best of our knowledge from the literature survey experimental results presented in this work show best results with supportive statistical inferenceObjective The aim of this work is to improve the quality of present state of art of diagnosis of Coronary Artery disease patients and prevent the delay and misdiagnosis of patients using proposed robust inference system The research work addresses the drawbacks inherited by the existing medical decision support systems and performance comparisons are based on specificity sensitivity accuracy and other measurable parameters The system consists of highly robust algorithms for training purpose and finally the output is given by the Bayes Voting ClassifierKernel Density Estimate Analyzer BVCKDE It addresses the problem of correctly predicting the onset of CADMethods The robust methods of treating CAD including sparse multinomial logistic regression Rotation forest ensemble with Support Vector Machines and Principal Components analysis Artificial Neural Networks CHAID decision tree Gaussian process Boosting methods and other machine learning models For relevant feature selection Rotation Forest is used consisting of Bayesian network and Genetic algorithm with PCA as projection filter and attribute ranking is obtained with ranker search algorithm Experiments are performed with confidence level of 95 and 99 along with Corrected Ttests two tailedResults The results obtained in this work are better and compared with the existing literatures available on the same problem The highest accuracy obtained by AdaBoost algorithm is 97619 To the best of our knowledge from the literature survey experimental results presented in this work show best results with supportive statistical inference Research has shown that several noninvasive features are highly correlated and this finding is confirmed in this study using factor analysisConclusion We have conducted the experiments using robust and reliable machine learning models choosing steep parameters for distinguishing CAD patients from healthy individuals based on the non invasive measurable features It paves a new path for the telemonitoring of Coronary heart disease patients based on noninvasive measurable features and can be extended over other diseases as a future work This work shows a high degree of software reliability and quality of the medical inference system
Keywords: