What Is Fetal Heart Rate?

Fetal heart rate is a physiological measurement that quantifies the number of cardiac contractions per minute in a developing fetus, typically ranging from 110 to 160 beats per minute in a healthy term pregnancy. It serves as the primary clinical indicator of fetal well-being during pregnancy and labor, because the autonomic nervous system regulates cardiac rhythm in direct response to oxygenation, activity, and neurological maturity. Persistent deceleration below 110 bpm or sustained elevation above 160 bpm are recognized patterns of concern that prompt clinical intervention.

The acquisition and interpretation of fetal heart rate signals sit at the intersection of biomedical signal processing, obstetric medicine, and sensor engineering. The goal is to distinguish physiological patterns from pathological ones accurately enough to guide care decisions without causing unnecessary interventions.

Signal Acquisition Methods

Cardiotocography is the standard clinical method for measuring fetal heart rate. An ultrasound transducer placed on the maternal abdomen uses Doppler shift to detect the movement of fetal heart valves, and a second transducer records uterine contractions simultaneously. The combined trace is the cardiotocograph, or CTG. An alternative approach, direct fetal electrocardiography, attaches a spiral electrode to the fetal scalp during labor to record the electrical signal closer to its source, improving signal-to-noise ratio at the cost of invasiveness. Non-invasive fetal ECG extraction via abdominal surface electrodes has become an active research direction, using adaptive filtering and blind source separation to remove the maternal ECG component and recover the weaker fetal signal.

Feature Extraction and Variability Analysis

The instantaneous heart rate derived from beat-to-beat intervals carries diagnostically important information beyond the mean rate. Fetal heart rate variability, the fluctuation of successive interbeat intervals, reflects the maturity and integrity of the autonomic nervous system. Linear analysis methods include computation of the mean, standard deviation of normal-to-normal intervals (SDNN), and power spectral density from autoregressive models. Nonlinear methods, such as approximate entropy, sample entropy, and detrended fluctuation analysis, capture complexity in the signal that linear statistics miss. A review of techniques for processing fetal heart rate signals catalogues how these features are extracted from both CTG and direct ECG sources.

Automated Interpretation

Manual CTG interpretation carries a well-documented problem of high inter-observer disagreement, a limitation that motivates computerized analysis. Automated systems apply rule-based classifiers, support vector machines, and more recently deep neural networks to CTG traces, assigning probability scores to defined pattern categories such as baseline bradycardia, late decelerations, and reduced variability. IEEE research on improving CTG monitoring through advanced signal processing has shown that multiparametric feature sets, combining morphological and spectral characteristics, outperform single-feature classifiers in detecting fetal compromise. Clinical decision support tools integrating such algorithms are in active evaluation across obstetric centers.

Applications

Fetal heart rate measurement and analysis have applications across several domains, including:

• Intrapartum monitoring to detect fetal distress during labor and guide delivery decisions
• Antepartum non-stress testing to evaluate fetal well-being in high-risk pregnancies
• Long-term ambulatory monitoring for pregnancies complicated by maternal diabetes or hypertension
• Wearable sensor development for home-based prenatal surveillance
• Machine learning benchmarking using publicly available fetal ECG datasets for algorithm validation