Information Theory to Probe Intrapartum Fetal Heart Rate Dynamics

Abstract

Intrapartum fetal heart rate (FHR) monitoring constitutes a reference tool in clinical practice to assess the baby's health status and to detect fetal acidosis. It is usually analyzed by visual inspection grounded on FIGO criteria. Characterization of intrapartum fetal heart rate temporal dynamics remains a challenging task and continuously receives academic research efforts. Complexity measures, often implemented with tools referred to as approximate entropy (ApEn) or sample entropy (SampEn), have regularly been reported as significant features for intrapartum FHR analysis. We explore how information theory, and especially auto-mutual information (AMI), is connected to ApEn and SampEn and can be used to probe FHR dynamics. Applied to a large (1404 subjects) and documented database of FHR data, collected in a French academic hospital, it is shown that (i) auto-mutual information outperforms ApEn and SampEn for acidosis detection in the first stage of labor and continues to yield the best performance in the second stage; (ii) Shannon entropy increases as labor progresses and is always much larger in the second stage; (iii) babies suffering from fetal acidosis additionally show more structured temporal dynamics than healthy ones and that this progressive structuration can be used for early acidosis detection.

Probing High-Order Dependencies With Information Theory

“…ii) Entropy rate h (m,τ ) mixes up marginal distribution features (variance and pointwise transform F ), hence static properties with dynamical properties and dependencies of any order, as indeed Σ (m) X depends jointly on c Y and on F , hence on the joint distribution of y (m,τ ) t . The fact that h (m,τ ) gather both static and dynamics aspects of time series likely explains why it has been observed in many applications to be a often discriminant feature, compared to other features [40]. However, the entanglement of static and dynamic properties may also be considered a drawback for the analysis, characterization and understanding of data properties.…”

Information Theory for Non-Stationary Processes with Stationary Increments

“…We then use the scaling law (19) for a = 1/t to relate the joint probability at a given time t to the joint probability at unit-time t = 1, which leads to:…”

Independent Analysis of Decelerations and Resting Periods through CEEMDAN and Spectral-Based Feature Extraction Improves Cardiotocographic Assessment

“…Several techniques have been proposed to study the nonlinear characteristics involved in the FHR signal in relation to fetal health status. In this context, mutual information (MI) has been employed to design efficient features for FHR signal analysis [33] and study the UC and FHR coupling [34]. Multivariate analysis based on linear and nonlinear features has been proposed to discriminate between normal and intrauterine growth-restricted fetuses [35].…”