Classification Methodology of CVD with Localized Features Analysis Using Phase Space Reconstruction Targeting Personalized Remote Health Monitoring

Naresh Vemishetty,Amit Acharyya,Shivteja Ayyagari,Soumya Jana,Saptarshi Das,Paolo Emilio Puddu,Koushik Maharatna

doi:10.22489/cinc.2016.126-512

Abstract

This paper introduces the classification methodology of Cardiovascular Disease (CVD) with localized feature analysis using Phase Space Reconstruction (PSR) technique targeting personalized health care. The proposed classification methodology uses a few localized features (QRS interval and PR interval) of individual Electrocardiogram (ECG) beats from the Feature Extraction (FE) block and detects the desynchronization in the given intervals after applying the PSR technique. Considering the QRS interval, if any notch is present in the QRS complex, then the corresponding contour will appear and the variation in the box count indicating a notch in the QRS complex. Likewise, the contour and the disparity of box count due to the variation in the PR interval localized wave have been noticed using the proposed PSR technique. ECG database from the Physionet (MIT-BIH and PTBDB) has been used to verify the proposed analysis on localized features using proposed PSR and has enabled us to classify the various abnormalities like fragmented QRS complexes, myocardial infarction, ventricular arrhythmia and atrial fibrillation. The design have been successfully tested for diagnosing various disorders with 98% accuracy on all the specified abnormal databases.

Highlights

According to the survey conducted by World Health Organization (WHO), Cardiovascular Diseases (CVD) are the primary drivers for the high mortality rate all over the world among all the non-communicable diseases [1]
Phase Space Reconstruction (PSR) technique has shown prospect of detection of Ventricular Arrhythmias (VA) [3] when analyzed statistics on large dataset of many samples. Since it works on several stored PQRST complexes globally there is a high chance of miss classification of important and interesting diagnostic features of the individual ECG beats consisting of fragmented QRS complexes or missing of P wave or any desynchronisation in the individual ECG beats
The methodology has been successfully tested for diagnosing various ECG abnormalities like fragmented QRS complex, atrial fibrillation, myocardial infarction and ventricular arrhythmia with 98 % of accuracy on all the subjects

Summary

Introduction

According to the survey conducted by World Health Organization (WHO), Cardiovascular Diseases (CVD) are the primary drivers for the high mortality rate all over the world among all the non-communicable diseases [1]. In the PSR technique, the original signal and the delayed version of it is plotted to get the dynamic system’s trajectory. PSR technique has shown prospect of detection of Ventricular Arrhythmias (VA) [3] when analyzed statistics on large dataset of many samples. Since it works on several stored PQRST complexes globally there is a high chance of miss classification of important and interesting diagnostic features of the individual ECG beats consisting of fragmented QRS complexes or missing of P wave or any desynchronisation in the individual ECG beats

Methods

Results

Conclusion