Detection of Atrial Fibrillation Episodes based on 3D Algebraic Relationships between Cardiac Intervals

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Atrial fibrillation (AF) is a common cardiac arrhythmia requiring accurate detection.
• Electrocardiogram (ECG) parameters like JT, QRS, and RR intervals are crucial for cardiac analysis.
• Existing methods for AF detection may benefit from enhanced analytical frameworks.

Purpose of the Study:

• To employ perfect matrices of Lagrange differences for detecting atrial fibrillation episodes.
• To investigate the sensitivity of algebraic relationships between cardiac intervals using extended matrix dimensions.
• To develop a decision support system for classifying individuals with potential AF indications.

Main Methods:

• Utilized three ECG parameters: JT interval, QRS interval, and RR interval.
• Applied the concept of perfect matrices of Lagrange differences, extending dimensions from two to three.
• Developed a decision support system incorporating statistical algorithms, probability distribution graphs, and semi-gauge indicator techniques for classification.

Main Results:

• Demonstrated increased sensitivity of algebraic relationships between cardiac intervals with a three-dimensional matrix.
• Established a baseline dataset for supervised classification.
• Successfully visualized the categorization of new candidates using probability graphs and semi-gauge indicators.

Conclusions:

• The study validates the sensitivity of perfect matrices of Lagrange differences for AF detection.
• A robust baseline dataset for supervised classification was established.
• The developed framework facilitates the classification of new candidates for early AF detection.