Automatic detection of uterine contractions before and during labor using EHG: A systematic review.
Giulia Acquaviva, Irene S Lensen, Elisabetta Peri et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review evaluated automatic detection methods for uterine contractions using electrohysterography (EHG) across 50 studies. It found that adaptive thresholding methods, particularly peak detection and RMS-envelope extraction, achieved high performance. However, Machine Learning and Deep Learning approaches require large annotated datasets for robustness, which may limit their clinical use.
Key findings
- Adaptive thresholding methods showed high performance in detecting uterine contractions.
- Machine Learning and Deep Learning techniques require large annotated datasets for effective training.
- The review identified significant heterogeneity across studies, limiting direct quantitative comparisons.
- Potential biases were noted due to lack of blinding in signal analysis.
Why it matters
Understanding the effectiveness of EHG-based contraction detection methods is crucial for improving monitoring during pregnancy. Enhanced detection could lead to timely interventions for preterm birth and other complications, ultimately improving maternal and fetal outcomes.
What to keep in mind
The review noted limitations such as the absence of a reliable reference for validating contractions and restrictive patient selection criteria.