Article

Prediction of countershock success using single features from multiple ventricular fibrillation frequency bands and feature combinations using neural networks

Department of Anaesthesiology and Critical Care Medicine, Innsbruck Medical University, Innsbruck, Austria.
Resuscitation (Impact Factor: 3.96). 06/2007; 73(2):253-63. DOI: 10.1016/j.resuscitation.2006.10.002
Source: PubMed

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.

0 Bookmarks
 · 
84 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: -The study aimed to validate the ability of amplitude spectrum area (AMSA) to predict defibrillation (DF) success and long-term survival in a large population of out-of-hospital cardiac arrests (CA). -ECG recorded by automated external defibrillators from different manufactures were obtained from CAs occurring in 8 city-areas. A database, including 2.447 DFs from 1.050 patients, was used as derivation group, while an additional database, including 1.381 DFs from 567 patients, served as validation. A 2-sec ECG window before DF was analyzed and AMSA calculated. Univariable, multivariable regression analyses, and area under the receiver operating characteristic (ROC) curve were used for associations between AMSA and study endpoints: DF success, sustained-ROSC, and long-term survival. Among the 2.447 DFs of the derivation database, 26.2% were successful. AMSA was significantly higher prior to a successful DF than a failing one (13±5 vs. 6.8±3.5 mV-Hz) and was an independent predictor of DF success (OR 1.33, 95%CI 1.20-1.37) and sustained-ROSC (OR 1.22, 95%CI 1.17-1.26). Area under the ROC curve for DF success prediction was 0.86 (95%CI 0.85-0.88). AMSA was also significantly associated with long-term survival. The following AMSA thresholds were identified: 15.5 mV-Hz for DF success, and 6.5 mV-Hz for DF failure. In the validation database, AMSA ≥15.5 mV-Hz had a positive predictive value (PV) of 84%, while AMSA ≤6.5 mV-Hz had a negative PV of 98%. -In this large derivation-validation study, AMSA was validated as an accurate predictor of DF success. AMSA appeared also as a predictor of long-term survival.
    Circulation 12/2014; DOI:10.1161/CIRCULATIONAHA.114.010989 · 14.95 Impact Factor
  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Previous investigations of out-of-hospital cardiac arrest (OHCA) have shown that the waveform characteristic amplitude spectral area (AMSA) can predict successful defibrillation and return of spontaneous circulation (ROSC) but has not been studied previously for survival. Objectives To determine whether AMSA computed from the ventricular fibrillation (VF) waveform is associated with pre-hospital ROSC, hospital admission, and hospital discharge. Methods Adults with witnessed OHCA and an initial rhythm of VF from an Utstein style database were studied. AMSA was measured prior to each shock and averaged for each subject (AMSA-avg). Factors such as age, sex, number of shocks, time from dispatch to monitor/defibrillator application, first shock AMSA, and AMSA-avg that could predict pre-hospital ROSC, hospital admission, and hospital discharge were analyzed by logistic regression. Results Eighty-nine subjects (mean age 62 ± 15 years) with a total of 286 shocks were analyzed. AMSA-avg was associated with pre-hospital ROSC (p = 0.003); a threshold of 20.9 mV-Hz had a 95% sensitivity and a 43.4% specificity. Additionally, AMSA-avg was associated with hospital admission (p < 0.001); a threshold of 21 mV-Hz had a 95% sensitivity and a 54% specificity and with hospital discharge (p < 0.001); a threshold of 25.6 mV-Hz had a 95% sensitivity and a 53% specificity. First-shock AMSA was also predictive of pre-hospital ROSC, hospital admission, and discharge. Time from dispatch to monitor/defibrillator application was associated with hospital admission (p = 0.034) but not pre-hospital ROSC or hospital discharge. Conclusions AMSA is highly associated with pre-hospital ROSC, survival to hospital admission, and hospital discharge in witnessed VF OHCA. Future studies are needed to determine whether AMSA computed during resuscitation can identify patients for whom continuing current resuscitation efforts would likely be futile.
    Journal of the American College of Cardiology 09/2014; 64(13):1362–1369. DOI:10.1016/j.jacc.2014.06.1196 · 15.34 Impact Factor