Automatic analysis of cardiac repolarization morphology using Gaussian mesa function modeling.

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Electrocardiogram (ECG) analysis is crucial for diagnosing cardiac conditions.
• Automated waveform analysis can improve diagnostic accuracy and efficiency.
• Identifying subtle repolarization abnormalities is challenging with traditional methods.

Purpose of the Study:

• To develop and validate a novel, fully automated method for wave identification and extraction from ECGs.
• To analyze repolarization waveform parameters using Gaussian mesa functions (GMFs).
• To assess the method's efficacy in distinguishing drug-induced changes and genetic predispositions like long QT syndrome (LQTS).

Main Methods:

• A machine-learning algorithm combined with Gaussian mesa functions (GMFs) for ECG waveform decomposition.
• Generalized orthogonal forward regression for GMF identification and feature extraction.
• Analysis of repolarization parameters in healthy subjects under sotalol administration and in LQTS patients (LQT1 and LQT2).

Main Results:

• The automated method demonstrated significant differences in repolarization parameters between baseline and sotalol-induced changes.
• GMF-based parameters were significantly altered in LQTS patients compared to healthy controls.
• T-wave width and repolarization descending phase were notably prolonged in LQT2 patients versus LQT1.

Conclusions:

• The novel automated ECG analysis method effectively identifies and quantifies repolarization abnormalities.
• This approach shows promise for diagnosing drug effects and genetic cardiac channelopathies like LQTS.
• GMF-based parameter analysis provides valuable insights into repolarization dynamics.