762: Remote labour detection using maternal electrohysterography and heart rate data acquired with a wearable sensor

Abstract

Labour detection may be helpful in providing just in time care and avoiding unnecessary antenatal visits. Given specific patterns in physiological data such as electrohysterography (EHG) and heart rate (HR) highlighted from previous literature in correspondence of uterine contractions, we sought to create a statistical model able to classify maternal recordings of EHG and HR data between labour recordings and non-labour recordings. EHG and HR data were collected on 59 pregnant women (26 in labour, defined as within 24 hours from delivery, and 33 not in labour) during the late stages of pregnancy (39.3 weeks mean gestational age at time of measurement) using a wearable sensor designed to be attached to the abdomen using an adhesive patch. We extracted time and frequency domain features from EHG and HR data, as stronger, sinusoidal pattern arise on both data streams in correspondence of uterine contractions during labour. Features were used as input to a statistical model, trained to recognize labour and non-labour recordings. The accuracy of the proposed model in classifying labour and non-labour recordings was evaluated using standard statistical techniques (leave one out cross validation). Accuracy in detecting labour was 89%, with 0.80 sensitivity and 0.94 specificity. Our labour detection model demonstrated high accuracy in classifying labour and non-labour recordings using EHG and HR data collected at late stage of pregnancy. This classifier may be useful to detect labour non-invasively, without the need for hospitalization or prenatal visit. Our model is tested prospectively in an independent cohort.View Large Image Figure ViewerDownload Hi-res image Download (PPT)