Prediction of countershock success using single features from multiple ventricular fibrillation frequency bands and feature combinations using neural networks.
ABSTRACT Targeted defibrillation therapy is needed to optimise survival chances of ventricular fibrillation (VF) patients, but at present VF analysis strategies to optimise defibrillation timing have insufficient predictive power. From 197 patients with in-hospital and out-of-hospital cardiac arrest, 770 electrocardiogram (ECG) recordings of countershock attempts were analysed. Preshock VF ECG features in the time and frequency domain were tested retrospectively for outcome prediction. Using band pass filters, the ECG spectrum was split into various frequency bands of 2-26 Hz bandwidth in the range of 0-26 Hz. Neural networks were used for single feature combinations to optimise prediction of countershock success. Areas under curves (AUC) of receiver operating characteristics (ROC) were used to estimate prediction power of single and combined features. The highest ROC AUC of 0.863 was reached by the median slope in the interval 10-22 Hz resulting in a sensitivity of 95% and a specificity of 50%. The best specificity of 55% at the 95% sensitivity level was reached by power spectrum analysis (PSA) in the 6-26 Hz interval. Neural networks combining single predictive features were unable to increase outcome prediction. Using frequency band segmentation of human VF ECG, several single predictive features with high ROC AUC>0.840 were identified. Combining these single predictive features using neural networks did not further improve outcome prediction in human VF data. This may indicate that various simple VF features, such as median slope already reach the maximum prediction power extractable from VF ECG.
- SourceAvailable from: Trygve Eftestøl[show abstract] [hide abstract]
ABSTRACT: We evaluated ventricular fibrillation frequency and amplitude variables to predict successful countershock, defined as pulse-generating electrical activity. We also elucidated whether bystander cardiopulmonary resuscitation (CPR) influences these electrocardiogram (ECG) variables. In 89 patients with out-of-hospital cardiac arrest, ECG recordings of 594 countershock attempts were collected and analyzed retrospectively. By using fast Fourier transformation analysis of the ventricular fibrillation ECG signal in the frequency range 0.333-15 Hz (median [range]), median frequency, dominant frequency, spectral edge frequency, and amplitude were as follows: 4.4 (2.4-7.5) Hz, 4.0 (0.7-7.0) Hz, 7.7 (3.7-13.7) Hz, and 0.94 (0.24-1.95) mV, respectively, before successful countershock (n = 59). These values were 3.8 (0.8-7.7) Hz (P = 0.0002), 3.0 (0.3-9.7) Hz (P < 0.0001), 7.3 (2.0-14.0) Hz (P < 0.05), and 0.53 (0.03-3.03) mV (P < 0.0001), respectively, before unsuccessful countershock (n = 535). In patients in whom bystander CPR was performed (n = 51), ventricular fibrillation frequency and amplitude before the first defibrillation attempt were higher than in patients without bystander CPR (n = 38) (median frequency, 4.4 [2.4-7.5] vs 3.7 [1.8-5.3] Hz, P < 0.0001; dominant frequency, 3.8 [0.9-7.7] vs 2.6 [0.8-5.9] Hz, P < 0.0001; spectral edge frequency, 8.4 [4.8-12.9] vs 7.2 [3.9-12.1] Hz, P < 0.05; amplitude, 0.79 [0.06-4.72] vs 0.67 [0.16-2.29] mV, P = 0.0647). Receiver operating characteristic curves demonstrate that successful countershocks will be best discriminated from unsuccessful countershocks by ventricular fibrillation amplitude (3000-ms epoch). At 73% sensitivity, a specificity of 67% was obtained with this variable. IMPLICATIONS: In patients with out-of-hospital cardiac arrest, successful countershocks will be best discriminated from unsuccessful countershocks by ventricular fibrillation amplitude (3000-ms epoch). At 73% sensitivity, a specificity of 67% was obtained with this variable.Anesthesia & Analgesia 01/2002; 93(6):1428-33, table of contents. · 3.30 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: To determine whether there is information in the human ventricular fibrillation (VF) ECG signal that is predictive of successful countershock. We carried out a retrospective analysis of ECG signals recorded during out-of-hospital treatment of adult patients in VF. Four parameters--centroid frequency (FC), peak power frequency (FP), average segment amplitude (SA), and average wave amplitude (WA)--were extracted from the recorded ECG signal immediately before each countershock and compared with countershock outcome. The outcome of each countershock (total, 128 countershocks) administered to 55 patients in VF was determined from available emergency medical services data sheets and time-domain ECG signal and voice recordings. The original 4-second time-domain ECG segment immediately before the countershock was used to extract SA and WA. The 4-second ECG segment immediately before each countershock was transformed into the frequency domain by means of Fourier analysis, and the parameters FC and FP were extracted from the result. These parameters were compared with countershock outcome by means of Kolmogrov-Smirnov analysis. Sensitivity and specificity of these parameters, as well as receiver operating characteristic curves, were constructed. FC was statistically higher for successful countershocks (FC, 5.48 +/- .67 Hz) than for successful countershocks (FC, 4.85 +/- 1.16 Hz; P=.012). We found no statistical difference for FP (P=.066), SA (P=.549), and WA (P =.337). FP and FC, when used in combination and in certain ranges (3.5 Hz < or = FP < or = 7.75 Hz and 3.86 Hz < or = FC < or = 6.12 Hz) had a sensitivity of 100% and a specificity of 47.1% in predicting successful countershock. The probabilities of predicting countershock outcome for FC, FP, SA, and WA were .72, .70, .52, and .53, respectively. FC and FP are predictive of countershock outcome for patients in VF and hold the potential to guide therapy during cardiac arrest.Annals of Emergency Medicine 02/1996; 27(2):184-8. · 4.29 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Cardiopulmonary resuscitation (CPR) creates artifacts on the ECG and, with automated defibrillators, a pause in CPR is mandatory during rhythm analysis. The rate of return of spontaneous circulation (ROSC) is reduced with increased duration of this hands-off interval in rats. We analyzed whether similar hands-off intervals in humans with ventricular fibrillation causes changes in the ECG predicting a lower probability of ROSC. The probability of ROSC after a shock was continually determined from ECG signal characteristics for up to 20 seconds of 634 such hands-off intervals in patients with ventricular fibrillation. In hands-off intervals with an initially high (40% to 100%) or median (25% to 40%) probability for ROSC, the probability was gradually reduced with time to a median of 8% to 11% after 20 seconds (P<0.001). In episodes with a low initial probability (0% to 25%; median, 5%), there was no further reduction with time. The interval between discontinuation of chest compressions and delivery of a shock should be kept as short as possible.Circulation 05/2002; 105(19):2270-3. · 15.20 Impact Factor