IMPROVING PHASE-RECTIFIED SIGNAL AVERAGING FOR FETAL HEART RATE ANALYSIS

Area of Science:

• Biomedical Engineering
• Neonatal Physiology
• Dynamical Systems Theory

Background:

• Low umbilical artery pH indicates neonatal acidosis, increasing complication risks.
• Phase-rectified signal averaging (PRSA) shows promise in fetal heart rate analysis but is parameter-sensitive.
• Current PRSA parameter selection relies on empirical methods or grid searches.

Purpose of the Study:

• To investigate the phase-rectified signal averaging (PRSA) method using dynamical systems theory.
• To reveal the connection between state space reconstruction and PRSA.
• To introduce a novel feature for improved dynamical system characterization in fetal heart rate analysis.

Main Methods:

• Analysis of PRSA through the lens of dynamical systems theory.
• Exploration of the relationship between state space reconstruction and PRSA.
• Development and testing of a new feature for dynamical system characterization.
• Experimental validation on an open-access intrapartum Cardiotocography database.

Main Results:

• A novel intrinsic connection between state space reconstruction and PRSA was identified.
• The proposed feature demonstrates superior characterization of dynamical systems compared to PRSA.
• Experimental results show the new feature outperforms state-of-the-art PRSA in pH-based fetal heart rate analysis.

Conclusions:

• The new feature offers improved performance over PRSA for fetal heart rate analysis.
• This approach provides a more robust method for assessing neonatal acidosis risk.
• Dynamical systems theory offers valuable insights into signal processing techniques like PRSA.