Phase Space Reconstruction Based CVD Classifier Using Localized Features

Naresh Vemishetty,Saptarshi Das,Amit Acharyya,Ramya Lakshmi Gunukula,Paolo Emilio Puddu,Koushik Maharatna

doi:10.1038/s41598-019-51061-8

Abstract

This paper proposes a generalized Phase Space Reconstruction (PSR) based Cardiovascular Diseases (CVD) classification methodology by exploiting the localized features of the ECG. The proposed methodology first extracts the ECG localized features including PR interval, QRS complex, and QT interval from the continuous ECG waveform using features extraction logic, then the PSR technique is applied to get the phase portraits of all the localized features. Based on the cleanliness and contour of the phase portraits CVD classification will be done. This is first of its kind approach where the localized features of ECG are being taken into considerations unlike the state-of-art approaches, where the entire ECG beats have been considered. The proposed methodology is generic and can be extended to most of the CVD cases. It is verified on the PTBDB and IAFDB databases by taking the CVD including Atrial Fibrillation, Myocardial Infarction, Bundle Branch Block, Cardiomyopathy, Dysrhythmia, and Hypertrophy. The methodology has been tested on 65 patients’ data for the classification of abnormalities in PR interval, QRS complex, and QT interval. Based on the obtained statistical results, to detect the abnormality in PR interval, QRS complex and QT interval the Coefficient Variation (CV) should be greater than or equal to 0.1012, 0.083, 0.082 respectively with individual accuracy levels of 95.3%, 96.9%, and 98.5% respectively. To justify the clinical significance of the proposed methodology, the Confidence Interval (CI), the p-value using ANOVA have been computed. The p-value obtained is less than 0.05, and greater F-statistic values reveal the robust classification of CVD using localized features.

Highlights

Contour as opposed to the diseased cases where the contour will be chaotic in nature
Healthy ECG samples followed by unhealthy samples of these databases are taken as different arrays (ECG database)
The outcome from the Feature Extraction (FE) block of all these arrays give the localized features, applying Phase Space Reconstruction (PSR) technique on these features results in PSR images, Fig. 1 shows the distribution of box-count in each image corresponding to the localized features

Summary

Introduction

Contour as opposed to the diseased cases where the contour will be chaotic in nature. Here we have introduced the concept of localized features to mitigate all the aforementioned limitations These are the reasons for which the ECG frame based classification as per the reported literature have the shortcomings that is rectified in the proposed method. It is evident from the above statement that all the aforementioned diseases can be detected using the localized diagnostic features that perhaps would be given a miss if the entire ECG frame is considered. We have taken only PR interval and QRS complex as a part of preliminary study and able to detect the abnormalities based on the box-count distribution This has motivated us to do the proposed work where we achieved the overwhelming results as reported in this manuscript. Based on achieved statistical analysis (ANOVA, Confidence Interval) and diagnosis measures, we conclude that the proposed methodology can be extended to classify most of the CVD cases

Objectives

Methods

Results