Three-dimensional ventricular activation imaging by means of equivalent current source modeling and estimation

Area of Science:

• Biomedical Engineering
• Computational Cardiology
• Medical Imaging

Background:

• Noninvasive imaging of the 3-D ventricular activation sequence is crucial for understanding cardiac electrophysiology and diagnosing heart conditions.
• Current methods often lack the spatial resolution or accuracy to fully capture the complex electrical propagation within the ventricles.

Purpose of the Study:

• To present a novel electrocardiographic inverse method for reconstructing the 3-D ventricular activation sequence.
• To estimate the equivalent current density (ECD) throughout the myocardium to derive activation times.
• To validate the approach using computer simulations on a realistic heart model.

Main Methods:

• Utilized spatio-temporal coherence of ventricular excitation to derive activation time from ECD.
• Employed a weighted minimum norm approach with spatio-temporal regularization (Singular Value Decomposition) to estimate ECD from body surface potential maps (BSPM).
• Determined activation time by identifying the time point of maximum local ECD estimate.

Main Results:

• Accurate reconstruction of the simulated 3-D ventricular activation sequence was achieved (average correlation coefficient: 0.90).
• The method demonstrated a low relative error (0.19) in activation sequence reconstruction.
• The origin of ventricular excitation was localized with a high degree of accuracy (average localization error: 5.5 mm).

Conclusions:

• The proposed electrocardiographic inverse approach enables accurate noninvasive imaging of the 3-D ventricular activation sequence.
• The method shows significant potential for clinical applications in diagnosing and understanding cardiac arrhythmias and electrical abnormalities.
• This technique offers a promising tool for detailed analysis of cardiac electrical propagation patterns.