Multiscale adaptive basis function modeling of spatiotemporal vectorcardiogram signals.

Area of Science:

• Biomedical Engineering
• Computational Cardiology
• Signal Processing

Background:

• Mathematical modeling of cardiac electrical signals is crucial for simulating cardiac behavior and evaluating algorithms.
• Existing models face challenges in efficacy, robustness, and generality for characterizing spatio-temporal patterns.

Purpose of the Study:

• To present a multiscale adaptive basis function modeling approach for characterizing temporal and spatial behaviors of vectorcardiogram (VCG) signals.
• To evaluate the model's performance across different basis functions and cardiac conditions.

Main Methods:

• Developed a multiscale adaptive basis function model where parameters are estimated via projections of VCG waves onto nonlinear basis functions.
• Experimentally evaluated model performance using various basis functions (Gaussian, Mexican hat, customized wavelet, Hermitian wavelets) and cardiac conditions (healthy controls, myocardial infarctions).

Main Results:

• Multiway ANOVA indicated basis function choice and model complexity significantly impact performance, while cardiac conditions did not.
• A customized wavelet emerged as the optimal basis function for spatio-temporal VCG signal modeling.
• Model representations showed <5% relative error in QT intervals compared to real VCG signals for model complexity >10.

Conclusions:

• The proposed model effectively captures spatio-temporal cardiac electrical behaviors, including pathological conditions.
• This approach holds significant potential for applications in feature extraction, data compression, algorithm evaluation, and disease prognostics.