Cardiac arrhythmia detection using cross-sample entropy measure based on short and long RR interval series

Area of Science:

• Biomedical Signal Processing
• Cardiovascular Health
• Nonlinear Dynamics

Background:

• Accurate detection of cardiac arrhythmias like atrial fibrillation (AF) and congestive heart failure (CHF) remains a challenge in biomedical signal processing.
• Electrocardiogram (ECG) signal analysis, employing linear and nonlinear measures, is crucial for addressing these challenges.

Purpose of the Study:

• To introduce and evaluate a novel nonlinear technique, cross-sample entropy (CrossSampEn), for quantifying health status using two-series analysis.
• To compare the efficacy and consistency of CrossSampEn against the established single-series Sample entropy (SampEn) method for arrhythmia detection.

Main Methods:

• The study utilized ECG records from subjects with normal sinus rhythm, AF, and CHF.
• Cross-sample entropy (CrossSampEn) was applied to analyze the irregularity of same and different R-R interval series across various data lengths.
• The proposed algorithm was validated using simulated data and statistical analysis (One-way ANOVA).

Main Results:

• CrossSampEn demonstrated superior consistency compared to SampEn, notably avoiding undefined values for short data lengths.
• Statistical validation (One-way ANOVA) confirmed the algorithm's significance with a large F value and p < .0001.
• The method effectively quantified irregularity in both same and different R-R interval series.

Conclusions:

• Optimal health-status detection requires different RR interval series of ~1500 data points and same RR interval series of ~1000 data points, with embedded dimensions M=2 and threshold r=0.2.
• CrossSampEn offers a more consistent and reliable measure for arrhythmia detection compared to Sample entropy.
• The findings support CrossSampEn as a valuable tool in biomedical signal processing for cardiovascular health assessment.