[Research on conditional fluctuation characteristics of CHF heart rate variation]

Area of Science:

• Cardiology
• Biomedical Engineering
• Time Series Analysis

Background:

• Heart Rate Variation (HRV) analysis is crucial for understanding cardiac health.
• Previous research has explored HRV but often lacks sophisticated modeling for its dynamic fluctuations.
• Identifying distinct HRV patterns in conditions like congestive heart failure is clinically significant.

Purpose of the Study:

• To investigate the conditional fluctuation characteristics of Heart Rate Variation (HRV) series.
• To apply Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models for HRV analysis.
• To differentiate HRV patterns between patients with congestive heart failure and healthy individuals.

Main Methods:

• Utilized the PhysioNet ECG database for heart rate variation data.
• Applied various GARCH model family members, including ARCH(1) and TGARCH(1,1).
• Evaluated model performance based on fitting the changing rate of HRV series.

Main Results:

• Confirmed the existence of conditional fluctuation characteristics within HRV series.
• The Threshold GARCH (1,1) (TGARCH (1,1)) model demonstrated superior performance in fitting HRV rate changes.
• Significant differences in HRV patterns were observed between the congestive heart failure group and the normal group.

Conclusions:

• The TGARCH (1,1) model provides an effective method for analyzing HRV fluctuations.
• The study highlights distinct HRV characteristics associated with congestive heart failure.
• These findings offer a novel approach for HRV research and clinical diagnostic applications.