A robust fetal ECG detection method for abdominal recordings

Physiol Meas. 2007 Apr;28(4):373-88. doi: 10.1088/0967-3334/28/4/004. Epub 2007 Mar 7.

Abstract

In this paper, we propose a new method for FECG detection in abdominal recordings. The method consists of a sequential analysis approach, in which the a priori information about the interference signals is used for the detection of the FECG. Our method is evaluated on a set of 20 abdominal recordings from pregnant women with different gestational ages. Its performance in terms of fetal heart rate (FHR) detection success is compared with that of independent component analysis (ICA). The results show that our sequential estimation method outperforms ICA with a FHR detection rate of 85% versus 60% of ICA. The superior performance of our method is especially evident in recordings with a low signal-to-noise ratio (SNR). This indicates that our method is more robust than ICA for FECG detection.

MeSH terms

  • Abdomen*
  • Algorithms*
  • Artificial Intelligence*
  • Diagnosis, Computer-Assisted / methods*
  • Electrocardiography / methods*
  • Fetal Monitoring / methods*
  • Humans
  • Pattern Recognition, Automated / methods*
  • Principal Component Analysis
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Signal Processing, Computer-Assisted*