Comparison of Electrocardiogram between Dilated Cardiomyopathy and Ischemic Cardiomyopathy Based on Empirical Mode...

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Ischemic cardiomyopathy (ICM) and dilated cardiomyopathy (DCM) share clinical similarities but differ significantly in treatment and prognosis.
• Accurate early differentiation is vital for patient outcomes, yet the gold standard, coronary angiography, is invasive.
• Electrocardiogram (ECG) signal analysis using advanced signal processing techniques like variational mode decomposition (VMD) presents a potential non-invasive alternative.

Purpose of the Study:

• To investigate the efficacy of VMD combined with bispectral and nonlinear feature extraction from ECG signals for differentiating ICM from DCM.
• To compare the performance of VMD against empirical mode decomposition (EMD) in classifying these cardiac conditions.

Main Methods:

• ECG signals from 87 subjects (44 DCM, 43 ICM) were pre-processed, denoised, and divided into heartbeats.
• ECG signals were decomposed using both EMD and VMD, with five modes selected via correlation analysis.
• Bispectral and nonlinear features were extracted from the selected modes, followed by classification using five machine learning models.

Main Results:

• The proposed VMD-based method achieved a highest classification accuracy of 98.30% for distinguishing DCM and ICM.
• VMD consistently demonstrated superior performance over EMD across various modes, ECG leads, and classification algorithms.
• The study validates the effectiveness of VMD in ECG analysis for cardiac condition differentiation.

Conclusions:

• VMD-based ECG analysis provides a highly accurate and non-invasive approach for differentiating between ICM and DCM.
• This technique offers a promising alternative to invasive diagnostic methods, potentially improving patient management and outcomes.
• The superiority of VMD highlights its potential as a valuable tool in cardiovascular diagnostics.