Classification of electrocardiographic P-wave morphology

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• The P-wave on an electrocardiogram (ECG) represents normal atrial activity in the human heart.
• Intermittent atrial fibrillation can present with altered P-wave morphology, suggesting underlying atrial conduction defects.
• Distinguishing these abnormal P-waves from normal ones is crucial for accurate diagnosis and patient management.

Purpose of the Study:

• To develop and validate a novel method for discriminating between normal P-waves and those indicative of atrial conduction defects.
• To improve the diagnostic accuracy for identifying subtle P-wave abnormalities associated with intermittent atrial fibrillation.

Main Methods:

• Utilized impulse response analysis of P-wave signals.
• Employed linear discrimination techniques based on various extracted parameters.
• Classified 40 ECG recordings, comprising 20 normal P-waves and 20 abnormal P-waves.

Main Results:

• The developed classification method achieved a correct classification rate in 37 out of 40 recordings.
• Demonstrated high diagnostic performance with a sensitivity of 95% and a specificity of 90%.
• Successfully differentiated between normal and abnormal P-wave morphologies indicative of conduction issues.

Conclusions:

• The proposed method effectively discriminates between normal and abnormal P-waves, aiding in the detection of atrial conduction defects.
• This technique offers a promising tool for the early identification of patients with intermittent atrial fibrillation.
• Impulse response analysis combined with linear discrimination provides a robust approach for ECG-based cardiac diagnostics.