ECG-based Classification of Resuscitation Cardiac Rhythms for Retrospective Data Analysis

Abstract

There is a need to monitor the heart rhythm in resuscitation to improve treatment quality. Resuscitation rhythms are categorized into: ventricular tachycardia (VT), ventricular fibrillation (VF), pulseless electrical activity (PEA), asystole (AS), and pulse-generating rhythm (PR). Manual annotation of rhythms is time-consuming and infeasible for large datasets. Our objective was to develop ECG-based algorithms for the retrospective and automatic classification of resuscitation cardiac rhythms. Methods: The dataset consisted of 1631 3-second ECG segments with clinical rhythm annotations, obtained from 298 out-of-hospital cardiac arrest patients. 47 wavelet and time domain based features were computed from the ECG. Features were selected using a wrapper-based feature selection architecture. Classifiers based on Bayesian decision theory, k-nearest neighbor, k-local hyperplane distance nearest neighbor, artificial neural network (ANN), and ensemble of decision trees were studied.

Existing System

High-Temporal Resolution Algorithm, Feature Selection and Support Vector Machines.

Proposed System

This paper lays the groundwork for automatic resuscitation rhythm classification using only ECG, for use in the resuscitation data review process. This work is a comprehensive technical description of ECG based five class resuscitation rhythm classifiers. It thoroughly describes features derived from the wavelet analysis of the ECG, and combines them with classical shock advice algorithmic features. It introduces an improved feature selection scheme based on a nested cross-validation (CV) technique, and multiple experiments with different classification approaches. Finally, an algorithm is proposed using an optimal set of features and the best classifier.

Architecture

Nested cross-validation architecture