[Paper Review] Uterine muscle networks: Connectivity analysis of the EHG during pregnancy and Labor

Noujoud Nader, Mahmoud Hassan+4 more|arXiv (Cornell University)|Apr 10, 2019

Preterm Birth and Chorioamnionitis30 references4 citations

TL;DR

This study introduces a network-based analysis of abdominal electrohysterography (EHG) signals to characterize uterine electrical activity during pregnancy and labor. By applying graph theory to correlation-based connectivity matrices derived from 4x4 abdominal electrodes, the framework reveals denser, more synchronized networks during labor, achieving superior classification performance over conventional EHG features for detecting labor onset and preterm labor risk.

ABSTRACT

In this paper, we propose a new framework to analyze the electrical activity of the uterus recorded by electrohysterography (EHG), from abdominal electrodes (a grid of 4x4 electrodes) during pregnancy and labor. We evaluate the potential use of the synchronization between EHG signals in characterizing electrical activity of the uterus during pregnancy and labor. The complete processing pipeline consists of i) estimating the correlation between the different EHG signals, ii) quantifying the connectivity matrices using graph theory-based analysis and iii) testing the clinical impact of network measures in pregnancy monitoring and labor detection. We first compared several connectivity methods to compute the adjacency matrix represented as a graph of a set of nodes (electrodes) connected by edges (connectivity values). We then evaluated the performance of different graph measures in the classification of pregnancy and labor contractions (number of women=35). A comparison with the already existing parameters used in the state of the art of labor detection and preterm labor prediction was also performed. Results show higher performance of connectivity methods when combined with network measures. Denser graphs were observed during labor than during pregnancy. The network-based metrics showed the highest classification rate when compared to already existing features. This network-based approach can be used not only to characterize the propagation of the uterine contractions, but also may have high clinical impact in labor detection and likely in the prediction of premature labor.

Motivation & Objective

• To develop a network-based approach for analyzing uterine electrical activity using non-invasive EHG recordings.
• To evaluate the clinical utility of graph-theory metrics in distinguishing between pregnancy and labor states.
• To improve labor detection and preterm labor prediction by integrating connectivity measures with existing EHG parameters.
• To assess the performance of various connectivity methods and network metrics in classifying uterine contraction patterns.

Proposed method

• Electrohysterography (EHG) signals were recorded from a 4x4 grid of abdominal electrodes during pregnancy and labor.
• Cross-correlation between EHG signals from each electrode pair was computed to estimate functional connectivity.
• An adjacency matrix was constructed from correlation values, representing a network of electrodes (nodes) connected by edges (connectivity strength).
• Graph theory was applied to quantify network topology using metrics such as clustering coefficient, characteristic path length, and global efficiency.
• Multiple connectivity estimation methods were compared to identify the most robust approach for network construction.
• Classification performance of network metrics was evaluated using data from 35 participants to distinguish pregnancy from labor.

Experimental results

Research questions

• RQ1 Can network-based analysis of EHG signals improve the detection of uterine contractions during labor compared to traditional EHG features?
• RQ2 How does the topological structure of uterine muscle networks change between pregnancy and labor?
• RQ3 Which connectivity estimation method and network metric combination yields the highest classification accuracy for labor detection?
• RQ4 To what extent can network measures predict preterm labor onset?
• RQ5 How do network properties reflect the propagation dynamics of uterine electrical activity?

Key findings

• Denser and more synchronized uterine muscle networks were observed during labor compared to pregnancy, indicating increased functional connectivity.
• Network-based metrics achieved the highest classification rate among all evaluated features for distinguishing labor from pregnancy.
• The combination of correlation-based connectivity estimation with graph-theory metrics significantly outperformed conventional EHG parameters in classifying contraction states.
• Global efficiency and clustering coefficient showed strong discriminative power between pregnancy and labor phases.
• The proposed framework demonstrated high potential for non-invasive, real-time monitoring of uterine activity with clinical relevance for labor onset detection.
• The results suggest that network analysis can enhance the sensitivity and specificity of preterm labor prediction systems.