Computers in biology and medicine · 2016
Beat-to-beat T-wave alternans detection using the Ensemble Empirical Mode Decomposition method.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Hasan Muhammad A, Chauhan Vijay S, Krishnan Sridhar
Аннотация
BACKGROUND: T-wave alternans (TWA) is defined as a consistent variation in the repolarization morphology that repeats on every other beat. This study aimed to evaluate beat-to-beat TWA detection using the Ensemble EMD (EEMD) method.
METHOD: A total of 108 recordings of standard 12-lead ECGs of 69 healthy subjects (17 females, 42±18 years; 52 males, 40±13 years) and 39 cardiac-condition patients (ischemic cardiomyopathy; ICM and dilated cardiomyopathy; DCM) with left ventricular ejection fractions (LVEF) ≤40% were studied. We first determined the QT interval of ECG via a template matching algorithm. Then, beat-to-beat T-waves were extracted to quantify beat-to-beat TWA. The EEMD method was applied to the T-wave time series to decompose them into a set of intrinsic mode functions (IMFs). The instantaneous frequency was measured by performing the Hilbert transform on the selected IMF for extracting the features. Four different classifiers were applied to the extracted features to assess and classify the existence of TWA in the ECG signal.
RESULTS: In the simulation study, the global classifier worked better than the subject-based classifier for detecting alternans in the T-waves. In addition, the average accuracy and sensitivity for detecting TWA were greater than 80%. In the real Holter ECG data obtained from Toronto General Hospital, the Ensemble classifier had higher classification accuracy, 74%, than other classifiers and a positive predictive value of 100%.
CONCLUSION: In conclusion, the proposed Ensemble EMD method with Ensemble classifier can be utilized for detecting beat-to-beat TWA in the ECG signal.
METHOD: A total of 108 recordings of standard 12-lead ECGs of 69 healthy subjects (17 females, 42±18 years; 52 males, 40±13 years) and 39 cardiac-condition patients (ischemic cardiomyopathy; ICM and dilated cardiomyopathy; DCM) with left ventricular ejection fractions (LVEF) ≤40% were studied. We first determined the QT interval of ECG via a template matching algorithm. Then, beat-to-beat T-waves were extracted to quantify beat-to-beat TWA. The EEMD method was applied to the T-wave time series to decompose them into a set of intrinsic mode functions (IMFs). The instantaneous frequency was measured by performing the Hilbert transform on the selected IMF for extracting the features. Four different classifiers were applied to the extracted features to assess and classify the existence of TWA in the ECG signal.
RESULTS: In the simulation study, the global classifier worked better than the subject-based classifier for detecting alternans in the T-waves. In addition, the average accuracy and sensitivity for detecting TWA were greater than 80%. In the real Holter ECG data obtained from Toronto General Hospital, the Ensemble classifier had higher classification accuracy, 74%, than other classifiers and a positive predictive value of 100%.
CONCLUSION: In conclusion, the proposed Ensemble EMD method with Ensemble classifier can be utilized for detecting beat-to-beat TWA in the ECG signal.