T-wave vector alternans detection based on Holter ECG recordings.

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• T-wave alternans (TWA) detection is crucial for assessing cardiac risk.
• Current spectral analysis methods for TWA require controlled environments with elevated heart rates.
• Detecting low-amplitude TWA in noisy, ambulatory ECG recordings remains challenging.

Purpose of the Study:

• To develop a novel method for reliable T-wave alternans detection from 3-channel Holter ECG recordings during normal daily activities.
• To overcome the limitations of existing TWA detection methods that necessitate controlled test environments.
• To introduce a new approach named T-wave Vector Alternance (TWVA) for enhanced TWA identification.

Main Methods:

• Utilized singular value decomposition (SVD) for sensitive differentiation of T-wave morphology.
• Applied the SVD-based method to 3-channel Holter ECG recordings.
• Proposed the T-wave Vector Alternance (TWVA) terminology for SVD-detected TWA.

Main Results:

• The proposed TWVA method successfully detected T-wave alternans.
• The method demonstrated high sensitivity in differentiating T-wave morphology even under noisy recording conditions.
• Successful application in a small cohort including normal subjects and subjects with TWA.

Conclusions:

• The TWVA method offers a reliable approach for T-wave alternans detection in ambulatory 24-hour Holter ECG recordings.
• Singular value decomposition provides a robust tool for analyzing T-wave morphology in noisy signals.
• This new method has the potential to improve non-invasive cardiac risk stratification by enabling TWA detection during routine daily activities.