Development of a signal processing and feature extraction framework for the safe passage study

Abstract

This work describes a framework that was developed to export and analyze maternal heartrate and uterine activity from a Monica AN24 ECG recording device. 9478 Traces from 5356 patients were processed, the mean recording length was 47 minutes. The framework implemented additional signal cleaning algorithms and extracted accelerations, decelerations, baseline traces, commonly used heartrate variability parameters, phase rectified signal average wavelets and identified beats that were missed by the Monica detection algorithm.