Feasibility Analysis of ECG-Based pH Estimation for Asphyxia Detection in Neonates

Area of Science:

• Biomedical Engineering
• Neonatal Medicine
• Cardiology

Background:

• Birth asphyxia poses significant mortality and morbidity risks in neonates.
• Current monitoring methods like arterial blood gas analysis are invasive, while pulse oximetry can be inaccurate.
• There is a critical need for noninvasive, accurate, and continuous monitoring solutions for asphyxiated neonates.

Purpose of the Study:

• To identify prominent electrocardiogram (ECG) features correlated with pH levels for noninvasive monitoring of neonatal birth asphyxia.
• To evaluate the potential of specific ECG parameters in distinguishing between asphyxiated and non-asphyxiated neonates.

Main Methods:

• Utilized a dataset of 274 simultaneous ECG and pH recordings from neonates.
• Applied principal component analysis and the Pan-Tompkins algorithm for ECG feature extraction.
• Employed Kruskal-Wallis and Dunn-Šidák tests to analyze differences in ECG features between asphyxiated and non-asphyxiated groups.

Main Results:

• Significant differences (p < 0.05) were observed in multiple ECG features (T/QRS, T Amplitude, Tslope, HR, QT, QTc) based on pH estimation in asphyxiated neonates.
• Key ECG features demonstrated significant variations between the asphyxiated and non-asphyxiated infant groups.
• The study identified specific ECG parameters as potential indicators of birth asphyxia.

Conclusions:

• Specific ECG features show significant differences related to pH levels in neonates, indicating their utility in assessing birth asphyxia.
• These ECG-based metrics offer a promising avenue for noninvasive, accurate, and continuous monitoring of asphyxiated neonates.
• Further research into these ECG features could lead to improved clinical management of birth asphyxia.