Local Characteristic Decomposition as a Novel Approach to Predict Sudden Cardiac Death in Congestive Heart Failure Patients
محورهای موضوعی : BioElectricAli Dorostghol 1 , Adel Maghsoudpour 2 , Ali Ghaffari 3 , Mansour Nikkhah-Bahrami 4
1 - Department of Mechanical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
2 - Department of Mechanical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
3 - Department of Mechanical Engineering, K. N. Toosi University of Technology,
Tehran, Iran
4 - Department of Mechanical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
کلید واژه: Congestive heart failure, Heart Rate Variability, Sudden Cardiac Death, Multi Fractal Dimension,
چکیده مقاله :
The current study uses heart rate variability (HRV) signal processing to investigate changes in the multifractal dimension in congestive heart failure (CHF) patients and predict sudden cardiac death (SCD). In this regard, HRV signals are first extracted, and their four sub-signals are determined using the Local Characteristic Decomposition (LCD) method. In the next step, using the Teager Energy method, the instant amplitudes of each sub-signal obtained in the previous step are calculated; thus, new signals are generated based on these instant amplitudes. Employing multifractal detrended fluctuation analysis (MF-DFA), the modified fractal dimensions of each new signal are then obtained. With the t-test method, appropriate features are selected and input into the support vector machine (SVM) classifier. By detecting subtle changes in HRV signals, this method can detect SCD in CHF patients. The results indicate that the proposed algorithm can distinguish the signals of SCD subjects with an accuracy of 84.76% 26 minutes before the event. In addition, after passing each 5-minute interval, the proposed method can update and determine how much time is left before SCD occurs
The current study uses heart rate variability (HRV) signal processing to investigate changes in the multifractal dimension in congestive heart failure (CHF) patients and predict sudden cardiac death (SCD). In this regard, HRV signals are first extracted, and their four sub-signals are determined using the Local Characteristic Decomposition (LCD) method. In the next step, using the Teager Energy method, the instant amplitudes of each sub-signal obtained in the previous step are calculated; thus, new signals are generated based on these instant amplitudes. Employing multifractal detrended fluctuation analysis (MF-DFA), the modified fractal dimensions of each new signal are then obtained. With the t-test method, appropriate features are selected and input into the support vector machine (SVM) classifier. By detecting subtle changes in HRV signals, this method can detect SCD in CHF patients. The results indicate that the proposed algorithm can distinguish the signals of SCD subjects with an accuracy of 84.76% 26 minutes before the event. In addition, after passing each 5-minute interval, the proposed method can update and determine how much time is left before SCD occurs
