An enhanced method for detecting congestive heart failure - Automatic Classifier

Abstract

A number of studies demonstrated the relationship of HRV (Heart Rate Variability) measures. Over the past years, automatic classifier, based on several clinical & instrumental parameters have been proposed to support CHF assessment. Considering only the low level features will not fulfill the classification needs. In order to avoid the gap between low level i.e general causes for CHF & high level features i.e attribute retrieved from long term HRV & make a decision correctly proposed a classifier to individuate severity of CHF. The proposed classifier separates lower risk patients from higher risk ones, using standard long-term heart rate variability (HRV) measures. The method we used to develop the Automatic Classifier is Bayesian belief network Classifier. The Bayesian Belief Network Classifier has been used in several applications especially for medical diagnosis. The Bayesian Belief Network algorithm iteratively splits the dataset, according to a criterion that maximizes the separation of the data which will produce a tree-like decision.