Atrial wave detection algorithm for discovery of some rhythm abnormalities.

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Atrial flutter (AFL) and atrial fibrillation (AF) present ECG patterns similar to P-waves, necessitating accurate detection for diagnosis.
• Existing P-wave detection methods often rely on manually selected templates, which can be suboptimal.
• P-wave detection is crucial for identifying arrhythmias like AFL/AF and for diagnosing AV block.

Purpose of the Study:

• To develop and evaluate a novel algorithm for P-wave detection using a universal synthesized template and modified convolution.
• To assess the algorithm's efficacy in recognizing AFL/AF arrhythmias and its potential for AV block identification.
• To compare the performance of the new method against existing template-matching approaches.

Main Methods:

• Employs a universal synthesized template for P-wave detection, avoiding manual template selection.
• Utilizes a modified convolution technique, incorporating sample differences in the denominator to enhance true coincidences.
• Implements a two-step convolution process with adaptive thresholding and an acceleration mechanism based on preliminary convolution results.

Main Results:

• The algorithm demonstrates successful detection of AFL/AF arrhythmias, with missed or wrong P-wave detections occurring in a limited number of RR intervals.
• The developed method shows a mean delay of 2.7s for AF onset recognition and 5.5s for AF offset recognition over a 10-hour recording.
• The integrated QRS detector allows for potential application in AV block identification.

Conclusions:

• The algorithm is effective for AFL/AF recognition, except for the prediction of paroxysmal AF.
• The method provides a robust and efficient approach to P-wave detection, contributing to cardiac arrhythmia diagnosis.
• The algorithm's dual capability in P-wave and QRS detection offers a versatile tool for cardiovascular signal analysis.