Feasibility Analysis of ECG-based pH estimation for Asphyxia Detection
in Neonates
Abstract
Birth asphyxia a potential cause of death is also associated with acute,
and chronic morbidities. The traditional and immediate approach to
monitor birth asphyxia (i.e., arterial blood gas analysis) is highly
invasive and intermittent. Additionally, alternative non-invasive
approaches such as pulse oximeter can be problematic, due to the
possibility of false and erroneous measurements. Therefore, further
research is needed to explore alternative non-invasive and accurate
monitoring methods for asphyxiated neonates. This study aims to
investigate prominent ECG features based on pH estimation that could
potentially be used to explore non-invasive, accurate, and continuous
monitoring of asphyxiated neonates. The dataset contains 274 segments of
ECG and pH values recorded simultaneously. After pre-processing of data,
principal component analysis and Pan-Tompkins algorithm are used for
each segment, to determine the most significant ECG cycle, and to
compute the ECG features. Descriptive statistics are performed to
describe the main properties of the dataset. The Kruskal-Wallis
nonparametric test is then used to analyse differences between the two
groups. Finally, Dunn–Šidák post-hoc test is utilised for individual
comparison among mean ranks of all groups. This study showed that ECG
features (mainly QT, QTc, and Tslope/ T) based on pH estimation differed
significantly in asphyxiated neonates.