Real-time electrogram analysis for monitoring coronary blood flow during human ventricular fibrillation: Implications for CPR

Analysis of electrocardiogram pre-shock waveforms during ventricular fibrillation

“…AMSA is a spectral feature derived from the spectrum and is the product of the frequency and its amplitude [11]. CS is the scale that has the maximum energy [20]. LAC is one of the chaotic features, and is computed as the logarithm of the summation of the correlation with different lags [14].…”

“…LAC is one of the chaotic features, and is computed as the logarithm of the summation of the correlation with different lags [14]. SDW is a wavelet-based feature and is computed as the number of scales that have energy values equal to or higher than a set threshold [20]. These features were chosen for their diverse methodologies, ensuring a thorough comparison between methodologies.…”

“…The total classification accuracy (TotalAcc) was calculated as the summation of the samples correctly classified in each group (Class c ) versus the total number of samples (Samples Total ), and is given by Eq. (20).…”

“…Wavelet analysis has also been proposed on the ECG during VF [15] and markers in the wavelet domain, such as cardioversion outcome prediction (COP) [16] and wavelet entropy [17], were introduced. Our preliminary works have also explored wavelet features and arrived at a relative scale distribution width (SDW) feature in predicting shock outcomes [18,19], expanding the application of this feature to quantify CPR efficacy by indirectly measuring coronary flow during VF [20]. Although the aforementioned works did arrive at markers that correlated with shock success, there are no studies (to the authors' knowledge) that attempt to relate the mechanistic insights of underlying sources that maintain VF and their relation in characterizing VF for predicting shock success.…”

Deciphering the electrogram in ventricular fibrillation to extract physiological information

“…In this issue of Heart Rhythm, Umapathy et al 12 show how wavelet analysis overcomes this problem and can lead to meaningful insight extraction of VF signals. A wavelet is a mathematical function used to divide a signal into different scale components.…”