Implementation of NICHD diagnostic criteria for feature extraction and classification of fetal heart rate signals

We present a description of a computerized approach to interpretation of fetal heart rate (FHR) patterns that uses the 2008 Fetal Heart Rate Monitoring Guidelines published by the National Institute of Child Health and Human Development. The study addresses accurate feature extraction that gives programmatic meaning to heretofore visually adjudged morphological changes in FHR that may be indicative of compromised fetal status. We develop methods for estimation of baseline heart rate, detection of morphological changes such as decelerations, and a novel method for assessing FHR variability. A strong motivation for this effort was to imitate physician judgement by following the standardized guidelines as closely as possible. We demonstrate the accuracy of this expert system on a database of 30 traces that had been originally labeled by two independent physicians.