Preterm pregnancies are one of the leading causes of morbidity and mortality amongst children under the age of 5. This is a global issue and has been identified as an area requiring active research. The emphasis now is to identify and develop methods of predicting the likelihood of preterm birth. Accurately predicting the potential of a preterm birth helps to develop and improve proactive care and treatment strategies for this vulnerable population.

This paper uses physiological data from a group of patients in active labour. The dataset contains information about foetal heart rate (FHR) and maternal heart rate (MHR) for all patients and Electro hysterogram (EHG) recordings for the measurement of uterine contractions. For the physiological data analysis and associated signal processing, we utilise Deep Wavelet Scattering (DWS). This is an unsupervised decomposition and feature extraction method combining characteristics from Deep Learning Convolutions, as well as the classical Wavelet Transform, to observe and investigate the extent to which active preterm labour can be accurately identified from an acquired physiological signal. Additional machine learning algorithms are tested on the acquired physiological data to allow for the identification of optimal model architecture for this specific physiological data.