Related Experiment Video
Updated: Jun 5, 2026
08:01
In utero Measurement of Heart Rate in Mouse by Noninvasive M-mode Echocardiography
Published on: November 22, 2013
Antepartum fetal heart rate feature extraction and classification using empirical mode decomposition and support
Niranjana Krupa1, Mohd Ali, Edmond Zahedi
1Department of Electrical Electronic and Systems Engineering, Universiti Kebangsaan Malaysia, Bangi, Malaysia. niranjana.krupa@gmail.com
Biomedical Engineering Online
|January 20, 2011
Summary
This study introduces a novel method using empirical mode decomposition (EMD) and support vector machine (SVM) for fetal heart rate (FHR) analysis. The approach shows promise in classifying FHR recordings as normal or at risk, aiding fetal surveillance.
Area of Science:
- Obstetrics and Gynecology
- Biomedical Engineering
- Signal Processing
Background:
- Cardiotocography (CTG) is the primary method for fetal surveillance.
- Fetal heart rate (FHR) trace interpretation relies heavily on clinician expertise.
- Objective FHR analysis methods are needed to supplement visual assessment.
Purpose of the Study:
- To propose a new approach for FHR feature extraction using empirical mode decomposition (EMD).
- To classify FHR recordings as 'normal' or 'at risk' using EMD features and support vector machine (SVM).
Main Methods:
- FHR data from 15 subjects were recorded at 4 Hz.
- A dataset of 90 records (20-minute duration) was created and labeled by obstetricians.
- Standard deviations of EMD components were used as input features for an SVM classifier.
Main Results:
- Five-fold cross-validation on the training set yielded 86% accuracy and 94.8% geometric mean of sensitivity and specificity.
- The proposed method achieved an 81.5% geometric mean on the testing set.
- Kappa values for training and testing sets were .923 and .684, respectively.
Conclusions:
- The proposed EMD-based feature extraction and SVM classification is a promising method for FHR signal analysis.
- This approach offers a potential improvement for objective fetal surveillance.
- Further validation may enhance its clinical utility in identifying fetal risk.
