Asymmetric detrended fluctuation analysis in neonatal stress

Area of Science:

• Neonatal physiology
• Non-invasive monitoring techniques
• Biomedical signal processing

Background:

• Neonatal stress detection is crucial for well-being.
• Heart rate (HR) variability analysis offers insights into autonomic nervous system function.
• Existing methods may lack precision in differentiating stress responses.

Purpose of the Study:

• To investigate the utility of asymmetric detrended fluctuation analysis (ADFA) for detecting stress in newborns.
• To analyze the fractal structure of heart rate inter-beat intervals during stress.
• To differentiate between HR accelerations and decelerations using scaling exponents.

Main Methods:

• Forty healthy term newborns were studied.
• Two stress stimuli were applied: heel lance and simulated pressure.
• Asymmetric detrended fluctuation analysis (ADFA) was used to calculate scaling exponents (α⁺ for decelerations, α⁻ for accelerations).

Main Results:

• During stress, the acceleration scaling exponent (α⁻) significantly increased and exceeded the deceleration exponent (α⁺).
• ROC curve analysis showed good diagnostic performance for α⁻ (AUC 0.626-0.826).
• Combined use of α⁺ and α⁻ improved diagnostic performance (AUC 0.691-0.833), especially for shorter time series.

Conclusions:

• ADFA, particularly the acceleration scaling exponent, shows promise as a clinical tool for monitoring neonatal stress.
• This method provides a quantitative approach to understanding autonomic responses to stress in newborns.
• The findings suggest a potential for improved objective assessment of neonatal well-being during stressful events.