This article presents a methodology based on the mixture model to classify the real biomedical time series. The mixture model is shown to be an efficient probabilistic density estimation scheme aimed at approximating the posterior... more
This article presents a methodology based on the mixture model to classify the real biomedical time series. The mixture model is shown to be an efficient probabilistic density estimation scheme aimed at approximating the posterior probability distribution of a certain class of data. The approximation is conducted by employing a weighted mixture of a finite number of Gaussian kernels whose
Research Interests:
We report on an improved method for the prediction of the outcome from electric shock therapy for patients in ventricular fibrillation: the primary arrhythmia associated with sudden cardiac death. Our wavelet transform-based marker, COP... more
We report on an improved method for the prediction of the outcome from electric shock therapy for patients in ventricular fibrillation: the primary arrhythmia associated with sudden cardiac death. Our wavelet transform-based marker, COP (cardioversion outcome prediction), is compared to three other well-documented shock outcome predictors: median frequency (MF) of fibrillation, spectral energy (SE) and AMSA (amplitude spectrum analysis). Optimum specificities for sensitivities around 95% for the four reported methods are 63 ± 4% at 97 ± 2% (COP), 42 ± 15% at 90 ± 7% (MF), 12 ± 3% at 94 ± 5% (SE) and 56 ± 5% at 94 ± 5% (AMSA), with successful defibrillation defined as the rapid (<60 s) return of sustained (>30 s) spontaneous circulation. This marked increase in performance by COP at specificity values around 95%, required for implementation of the technique in practice, is achieved by its enhanced ability to partition pertinent information in the time-frequency plane. COP therefore provides an optimal index for the identification of patients for whom shocking would be futile and for whom an alternative therapy should be considered.
Research Interests:
Research Interests:
Research Interests:
Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, Europe...more
Research Interests:
Although modern defibrillators are nearly always successful in terminating ventricular fibrillation (VF), multiple defibrillation attempts are usually required to achieve return of spontaneous circulation (ROSC). This is potentially... more
Although modern defibrillators are nearly always successful in terminating ventricular fibrillation (VF), multiple defibrillation attempts are usually required to achieve return of spontaneous circulation (ROSC). This is potentially deleterious as cardiopulmonary resuscitation (CPR) must be discontinued during each defibrillation attempt which causes deterioration in the heart muscle and reduces the chance of ROSC from later defibrillation attempts. In this work defibrillation outcomes are predicted prior to electrical shocks using a neural network model to analyse VF time series in an attempt to avoid defibrillation attempts that do not result in ROSC. The 198 pre-shock VF ECG episodes from 83 cardiac arrest patients with defibrillation conversions to different outcomes were selected from the Oslo ambulance service database. A probabilistic neural network model was designed for training and testing with a cross validation method being used for the better generalisation performance. We achieved an accuracy of 75% in overall prediction with a sensitivity of 84% and a specificity of 65% using VF ECG time series of an order of 1 s in length. Pre-shock VF ECG time series can be classified according to the defibrillation conversion to a return of spontaneous circulation (ROSC) or No-ROSC.
Research Interests: Nonparametric Statistics, Time Series, Electrocardiography, Emergency Medical Services, Cardiac arrest, and 10 moreHumans, Outcome, Resuscitation, Clinical Sciences, Cardiopulmonary Resuscitation, Public health systems and services research, Probabilistic Neural Network, Area Under Curve, Ventricular Fibrillation, and Neural Network Model
Research Interests:
Research Interests:
Research Interests:
Research Interests:
ABSTRACT
Research Interests:
ABSTRACT
Research Interests:
Research Interests:
Research Interests:
Research Interests: Emergency Medical Services, Cardiac arrest, Prospective studies, Humans, Female, and 17 moreResuscitation, Blood sampling, Male, Ventilation, Risk factors, Clinical Sciences, Cardiopulmonary Resuscitation, Aged, Middle Aged, Oxygen, Public health systems and services research, Time Factors, Mechanical Ventilation, Risk Factors, Advanced Life Support, Pulmonary ventilation, and Tidal volume
Research Interests:
Research Interests:
European Resuscitation Council guidelines for the use of automated external defibrillators by EMS providers and first responders - A statement from the Early Defibrillation Task Force, with contributions from the Working Groups on Basic and Advanced Life Support, and approved by the Executive Com...more
Research Interests:
CPR creates artefacts on the ECG, and a pause in CPR is therefore mandatory during rhythm analysis. This hands-off interval is harmful to the already marginally circulated tissues during CPR, and if the artefacts could be removed by... more
CPR creates artefacts on the ECG, and a pause in CPR is therefore mandatory during rhythm analysis. This hands-off interval is harmful to the already marginally circulated tissues during CPR, and if the artefacts could be removed by filtering, the rhythm could be analyzed during ongoing CPR. Fixed coefficient filters used in animals cannot solve this problem in humans, due to overlapping frequency spectra for artefacts and VF signals. In the present study, we established a method for mixing CPR-artefacts (noise) from a pig with human VF (signal) at various signal-to-noise ratios (SNR) from -10 dB to +10 dB. We then developed a new methodology for removing CPR artefacts by applying a digital adaptive filter, and compared the results with this filter to that of a fixed coefficient filter. The results with the adaptive filter clearly outperformed the fixed coefficient filter for all SNR levels. At an original SNR of 0 dB, the restored SNRs were 9.0+/-0.7 dB versus 0.9+/-0.7 dB respectively (P&lt;0.0001).
Research Interests: Adaptive Filtering, Electrocardiography, Humans, Animals, Resuscitation, and 9 moreFeasibility Studies, Adaptive Filter, Clinical Sciences, Cardiopulmonary Resuscitation, Public health systems and services research, Swine, Sensitivity and Specificity, Signal to Noise Ratio, and Ventricular Fibrillation
Research Interests: Probability, Hemodynamics, Female, Animals, Resuscitation, and 15 moreMale, Clinical Sciences, Cardiopulmonary Resuscitation, Haemodynamics, Public health systems and services research, Analysis of Variance, Swine, Blood Flow, Coronary Circulation, Sensitivity and Specificity, Reference Values, Electric Impedance, Respiratory Mechanics, Ventricular Fibrillation, and Tidal volume
Research Interests:
Research Interests:
Research Interests: Algorithms, Computer Graphics, Ontology, Access To Information, Systems Integration, and 17 moreDatabase Management Systems, Emergency Medical Services, Cardiac arrest, Humans, Resuscitation, Standardisation, Medical Device, Clinical Sciences, Computer User Interface Design, Information Storage and Retrieval, Hospital Information Systems, Cardiopulmonary Resuscitation, Data representation, Information Dissemination, Public health systems and services research, Computer communication networks, and Automatic data processing
While pre-hospital factors related to outcome after out-of-hospital cardiac arrest (OHCA) are well known, little is known about possible in-hospitals factors related to outcome. Some in-hospital factors are associated with outcome in... more
While pre-hospital factors related to outcome after out-of-hospital cardiac arrest (OHCA) are well known, little is known about possible in-hospitals factors related to outcome. Some in-hospital factors are associated with outcome in terms of survival. An historical cohort observational study of all patients admitted to hospital with a spontaneous circulation after OHCA due to a cardiac cause in four different regions in Norway 1995-1999: Oslo, Akershus, Østfold and Stavanger. In Oslo, Akershus, Østfold and Stavanger 98, 84, 91 and 186 patients were included, respectively. Hospital mortality was higher in Oslo (66%) and Akershus (64%) than in Østfold (56%) and Stavanger (44%), P=0.002. By multivariate analysis the following pre-arrest and pre-hospital factors were associated with in-hospital survival: age &lt;or=71 years, better pre-arrest overall performance, a call-receipt-start CPR interval &lt;or=1 min, and no use of adrenaline (epinephrine). The in-hospital factors associated with survival were: no seizures, base excess &gt;-3.5 mmol l(-1), body temperature &lt;or=37.8 degrees C, and serum glucose &lt;or=10.6 mmol l(-1) 1-24 h after admittance with OR (95% CI) 2.72 (1.09-8.82, P=0.033), 1.12 (1.02-1.23, P=0.016), 2.67 (1.17-6.20, P=0.019) and 2.50 (1.11-5.65, P=0.028), respectively. Pre-arrest overall function, whether adrenaline was used, body temperature, the occurrence of hypotensive episodes, and the degree of metabolic acidosis differed between the four regions in parallel with the in-hospital survival rates. Both pre-arrest, pre- and in-hospital factors were associated with in-hospital survival after OCHA. It seems important also to report in-hospital factors in outcome studies of OCHA. The design of the study precludes a conclusion on causability.
Research Interests: Treatment Outcome, Comparative Study, Multivariate Analysis, Norway, Emergency Medical Services, and 20 moreHumans, Outcome, Female, Resuscitation, Male, Cohort Study, Temperature, Risk factors, Body Temperature, Clinical Sciences, Cardiopulmonary Resuscitation, Aged, Middle Aged, Public health systems and services research, Survival Rate, Metabolic Acidosis, Risk Factors, Hospital Mortality, Observational Study, and Cohort Studies
There is a need for robust, effective predictors of the outcome from shock for out-of-hospital cardiac arrest patients. Such technology would enable the emergency responder to provide a therapy tailored to the patient&#39;s needs.... more
There is a need for robust, effective predictors of the outcome from shock for out-of-hospital cardiac arrest patients. Such technology would enable the emergency responder to provide a therapy tailored to the patient&#39;s needs. Here we report our most recent findings while dwelling intentionally on the rationale behind the decisions taken during system development. Specifically, we illustrate the need for sensible data selection, fully cross-validated results and the care necessary when evaluating system performance. We analyze 878 pre-shock ECG traces, all of at least 10 s duration from 110 patients with cardiac arrest of cardiac aetiology. The continuous wavelet transform was applied to preshock segments of ECG trace. Time-frequency markers are extracted from the transform and a linear threshold derived from a training set to provide high sensitivity prediction of successful defibrillation. These systems are then evaluated on a withheld test set. All experiments are cross-validated. When compared to popular Fourier-based techniques our wavelet transform method, COP (Cardioversion Outcome Predictor), provides a 10-20% improvement in performance with values of 66 +/- 4 specificity at 95 +/- 4 sensitivity, 61 +/- 4 specificity at 97 +/- 2 sensitivity and 56 +/- 1 specificity at 98 +/- 2 sensitivity achieved for datasets limited to 3, 6, and 9 shocks per patient, respectively. Thus, the assessment of the wavelet marker was associated with a high specificity value at or above 95% sensitivity in comparison to previously reported methods. Therefore, COP could provide an optimal index for the identification of patients for whom shocking would be futile, and for whom an alternative therapy could be considered.
Research Interests:
Research Interests:
We report an improved method for the estimation of shock outcome prediction based on novel wavelet transform-based time-frequency methods. Wavelet-based peak frequency, energy, mean frequency, spectral flatness and a new entropy measure... more
We report an improved method for the estimation of shock outcome prediction based on novel wavelet transform-based time-frequency methods. Wavelet-based peak frequency, energy, mean frequency, spectral flatness and a new entropy measure were studied to predict shock outcome. Of these, the entropy measure provided optimal results with 60 +/- 6% specificity at 91 +/- 2% sensitivity achieved for the prediction of return of spontaneous circulation (ROSC). These results represent a major improvement in shock prediction in human ventricular fibrillation.
Research Interests:
Research Interests: Australia, Adolescent, Cardiac arrest, Prospective studies, Humans, and 24 moreChild, Europe, Text, Female, Resuscitation, Feasibility Studies, Male, Cohort Study, Infant, Design Rationale, Randomized Trial, Registries, Clinical Sciences, North America, Cardiopulmonary Resuscitation, Aged, Middle Aged, First record, Adult, Public health systems and services research, Randomized Controlled Trial, Ventricular Tachycardia, Ventricular Fibrillation, and Child preschool
Research Interests: Fractal Geometry, Algorithms, Nonlinear dynamics, Forecasting, Fractals, and 23 moreProbability, Electrocardiography, Sudden Death, Humans, Female, Animals, Resuscitation, Male, Angular velocity, Clinical Sciences, Cardiopulmonary Resuscitation, Phase Space, Public health systems and services research, Time Factors, Swine, Three Dimensional, Development Strategy, PROBABILITY DENSITY, Rotation, Sensitivity and Specificity, Predictive value of tests, Ventricular Fibrillation, and Scaling Exponent
Although modern defibrillators are nearly always successful in terminating ventricular fibrillation (VF), multiple defibrillation attempts are usually required to achieve return of spontaneous circulation (ROSC). This is potentially... more
Although modern defibrillators are nearly always successful in terminating ventricular fibrillation (VF), multiple defibrillation attempts are usually required to achieve return of spontaneous circulation (ROSC). This is potentially deleterious as cardiopulmonary resuscitation (CPR) must be discontinued during each defibrillation attempt which causes deterioration in the heart muscle and reduces the chance of ROSC from later defibrillation attempts. In this work defibrillation outcomes are predicted prior to electrical shocks using a neural network model to analyse VF time series in an attempt to avoid defibrillation attempts that do not result in ROSC. The 198 pre-shock VF ECG episodes from 83 cardiac arrest patients with defibrillation conversions to different outcomes were selected from the Oslo ambulance service database. A probabilistic neural network model was designed for training and testing with a cross validation method being used for the better generalisation performance. We achieved an accuracy of 75% in overall prediction with a sensitivity of 84% and a specificity of 65% using VF ECG time series of an order of 1 s in length. Pre-shock VF ECG time series can be classified according to the defibrillation conversion to a return of spontaneous circulation (ROSC) or No-ROSC.
Research Interests: Nonparametric Statistics, Time Series, Electrocardiography, Emergency Medical Services, Cardiac arrest, and 10 moreHumans, Outcome, Resuscitation, Clinical Sciences, Cardiopulmonary Resuscitation, Public health systems and services research, Probabilistic Neural Network, Area Under Curve, Ventricular Fibrillation, and Neural Network Model
Research Interests:
Research Interests: Documentation, Treatment Outcome, Adolescent, Emergency Medical Services, Humans, and 22 moreChild, Outcome, Europe, Female, Resuscitation, Male, Cohort Study, Infant, Data Collection, Clinical Sciences, Intensive Care Unit, Aged, Middle Aged, Adult, Public health systems and services research, Time Factors, Medical Records, System Evaluation, Wounds and Injuries, Trauma Centers, Injury Severity Score, and Child preschool
Research Interests:
Research Interests:
Research Interests:
The ability of an organism to withstand trauma is determined by the injury per se and inherent properties of the organism at the time of injury. We analyzed whether pre-injury morbidity scored on a four-level ordinal scale according to... more
The ability of an organism to withstand trauma is determined by the injury per se and inherent properties of the organism at the time of injury. We analyzed whether pre-injury morbidity scored on a four-level ordinal scale according to the American Society of Anesthesiologists Physical Status (ASA-PS) classification system predicts mortality after trauma. From a total of 3,773 prospectively collected patients (years 2000-2004), 3,728 patients were included. Main outcome measure was mortality 30 days after injury. The effect of pre-injury ASA-PS on mortality was assessed using linear logistic regression analysis, controlling for Revised Trauma Score (RTS), Injury Severity Score (ISS), and age. Mortality increased with increasing pre-injury ASA-PS, age, and ISS, and with decreasing RTS. Unadjusted mortality rates were 5.7% in ASA-PS 1, 12.3% in ASA-PS 2, and 26.4% in ASA-PS 3-4. This increasing mortality trend across pre-injury ASA-PS group was evident in nearly all categories of ISS, RTS, and age. Odds ratio for death was 1.76 (95% CI, 1.14-2.72) for pre-injury ASA-PS 2, and 2.25 (95% CI, 1.36-3.71) for ASA-PS 3-4 compared with for ASA-PS 1 and adjusted for ISS, RTS, and age. There were no interaction effects between pre-injury ASA-PS and the other variables. Pre-injury ASA-PS score was an independent predictor of mortality after trauma, also after adjusting for the major variables in the traditional TRISS (Trauma and Injury Severity Score) formula. Including pre-injury ASA-PS score might improve the predictive power of a survival prediction model without complicating it.