The first cry – a marker for oxygen deprivation preceding delivery?

Abstract

Background The acoustic characteristics of a newborn’s first cries remain largely unexplored. Although subjective cry assessment plays a role in neonatal diagnostics, objective acoustic analyses are largely lacking. As an important parameter of the newborn’s status, the acid-base balance is hypothesised to influence the quality of the first vocalisations and thus be measurable purely through acoustic analysis. Methods This thesis investigates the statistical relationship between the acoustic and sequential characteristics of newborn vocalisations in the first minutes after birth and the newborn’s acid-base status (pH value, base excess). The data is also used to train a machine learning model that predicts pH values just based on acoustic characteristics. Results Significant correlations have been found between single acoustic properties of neonatal vocalisations and newborn acid-base status. This was particularly evident in parameters such as loudness, jitter, and shimmer. In addition, the feasibility of a machine learning approach to automatically pH values based on acoustic parameters was shown. Conclusions This work provides a valuable foundation for the further development of neonatal diagnostics using acoustic analysis. For the first time, it has been demonstrated that changes in acid-base status also result in changes of the acoustic properties of newborn vocalisations. In the future, acoustic analysis could be a technically simple, non-invasive and inexpensive method for evaluating a newborn’s health status.

Datum der Bewilligung	2025
Originalsprache	Englisch
Gradverleihende Hochschule	Medizinische Universität Graz
Betreuer/-in	Florian Pokorny (Betreuer*in) & Markus Gugatschka (Mitbetreuer*in)