Detection of structural changes in tachogram series for the diagnosis of atrial fibrillation events

Area of Science:

• Cardiology
• Biomedical Engineering
• Statistical Signal Processing

Background:

• Atrial Fibrillation (AF) is a prevalent cardiac arrhythmia associated with increased mortality.
• Electrocardiographic signal analysis, particularly tachogram series, is crucial for AF detection and event timing.
• Existing methods may struggle with identifying AF event boundaries amidst irregular heart rate variability.

Purpose of the Study:

• To develop and validate a novel statistical method for precise identification of Atrial Fibrillation event onset and termination.
• To assess the performance of the proposed method using simulated and real patient data.
• To improve the detection of AF events, especially in cases with preceding or succeeding irregular heart rate patterns.

Main Methods:

• Utilized the order identification of Autoregressive Integrated Moving Average (ARIMA) models applied to RR time series.
• Characterized distinct phases of AF (pre-, during, post-AF) using ARIMA model parameters.
• Evaluated the method's efficacy through a simulation study and application to clinical patient data.

Main Results:

• The proposed statistical method demonstrated effectiveness in identifying structural changes in ARIMA models corresponding to AF events.
• The method successfully pinpointed the starting and ending points of Atrial Fibrillation episodes.
• Performance was validated on both simulated data and real-world patient datasets, showing robust detection capabilities.

Conclusions:

• The novel ARIMA model-based statistical approach offers a reliable tool for accurate Atrial Fibrillation event detection.
• This method enhances the ability to precisely determine the temporal boundaries of AF episodes, even in complex cardiac rhythm scenarios.
• The findings contribute to improved monitoring and management strategies for patients with Atrial Fibrillation.