Heart rate variability, multifractal multiscale patterns and their assessment criteria

Area of Science:

• Cardiology
• Physiology
• Complex Systems Analysis

Background:

• The central and autonomic nervous systems intricately regulate heart rate (HR) through complex, multifractal dynamics.
• Human heart rate variability (HRV) analysis is crucial for understanding cardiovascular health.
• Existing methods may not fully capture the complexity of HRV across various time scales.

Purpose of the Study:

• To analyze human heart rate variability (HRV) using nonlinear multiscale multifractal analysis (MMA).
• To investigate the effectiveness of MMA in distinguishing between healthy individuals and patients with various cardiovascular diseases.
• To explore MMA's potential for medical screening by analyzing low-frequency HRV components.

Main Methods:

• Employed nonlinear multiscale multifractal analysis (MMA) on night-time HRV recordings.
• Analyzed data from six patient groups: healthy, aortic valve stenosis, hypertrophic cardiomyopathy, atrial fibrillation, coronary artery disease, and congestive heart failure.
• Utilized a training data set (85%) and a test data set (15%) for model validation.

Main Results:

• MMA generated a unique Hurst surface for each medical condition, aiding differentiation.
• Achieved diagnostic accuracies ranging from 67% to 88% for specific diseases in the training set.
• Demonstrated an overall accuracy of 73% for the training set and 80% for the test set.

Conclusions:

• MMA provides valuable insights into HRV complexity, aiding cardiovascular disease diagnosis.
• The method shows potential as a novel screening tool, particularly for low-frequency HRV changes.
• Further research into the complete network moderating heart rhythm is warranted, considering MMA's findings.