Prediction of the success of cardiac resuscitation: a pattern classification approach based on ECG spectral and temporal features

Introduction

The ECG tracings recorded during a ventricular fibrillation (VF) using an automated external defibrillator (AED) contain useful information predictive of shock outcome. The focus is on the VF waveforms' morphology. The amplitude and the spectral properties of VF may predict the likelihood of successful defibrillation [1–4]. In almost all previous studies, the amplitude or the spectral properties of the ECG tracings have been singularly used. However, these approaches have led to methods lacking sufficient predictive power.

Methods

Five hundred patients with out-of-hospital cardiac arrest on arrival in an emergency room were examined. The rhythm was identified as VF and confirmed by two trained investigators. ECG data were stored in modules in digitized form over a period of 20 minutes and were analyzed retrospectively. ECG traces containing CPR artefacts were removed by digital filtering. Times of collapse, dispatch, scene arrival, CPR, and initial defibrillation were determined from dispatch records, recordings of arrest events, interviews with bystanders, and hospital records. The preshock VF waveform morphology was studied and different parameters of VF ECG signals were extracted. We then introduced a pattern classification machine that combines the amplitude and spectral features simultaneously.

Results

The use of the pattern classification machine which combines amplitude and spectral features of VF ECG signals shows an improved predictive power as compared with other methods.

Conclusions

This technique could help to determine which patients should receive shock first and which should receive a period of CPR prior to shock, thereby increasing the probability of survival. The potential impact of this research is high in the direction of generating a new methodology able to increase the probability of survival after a cardiac crisis.