Algebraic decomposition of the TU wave morphology patterns

Area of Science:

• Cardiovascular Physiology
• Biomathematics
• Electrocardiography

Background:

• The T wave on an electrocardiogram (ECG) reflects ventricular repolarization.
• T wave morphology is influenced by variations in action potential durations across ventricular regions.
• Existing methods for analyzing T wave morphology may lack detailed quantitative descriptors.

Purpose of the Study:

• To develop and validate a mathematical model for T wave morphology analysis.
• To represent T wave patterns as a summation of action potential-like functions.
• To characterize T wave morphology using a set of eight numerical parameters.

Main Methods:

• A mathematical model TU = S1 - S2 + L1 - L2 was formulated using Hill's equation for sigmoidal curves.
• Each component function was defined by amplitude (Amax) and duration (D95) parameters, resulting in eight total parameters.
• The model was validated against 170 digitized T wave complexes from normal subjects and patients with abnormal T waves (leads V2-V6).

Main Results:

• The model achieved high correlation coefficients (>= 0.99) between original T wave patterns and modeled data.
• Mean absolute differences between observed and modeled T wave values were consistently low for both normal and abnormal morphologies.
• The eight-parameter model effectively described and categorized diverse T wave patterns.

Conclusions:

• The developed mathematical model accurately represents T wave morphology based on underlying action potential differences.
• The eight numerical parameters provide a detailed and quantitative characterization of T wave patterns.
• This model offers a robust tool for analyzing and categorizing various T wave morphologies in electrocardiography.