Korean J Obstet Gynecol.  2005 Dec;48(12):2797-2806.

Fetal Well-being Evaluation by Determination of High Frequency Measurement Noise for Fetal Heart Rate Signal

Affiliations
  • 1Department of Obstetric and Gynecology, Medical College, Catholic University, Seoul, Korea. mamacare@catholic.ac.kr
  • 2Department of Physics, Pohang University of Science and Technology, Pohang, Korea.

Abstract


OBJECTIVE
This study was done to propose a new method of determining the cutoff frequency against measurement noise in fetal heart rate signal and to analyze the discriminating value for normal fetus (pH>7.15) and abnormal fetus (pH< or =7.15) by nonlinear method.
METHODS
The patients were classified by umbilical artery pH (7.15) or base excess (-10 mEq/L). After removing the high frequency noise components at the cutoff frequency determined in this study, we calculated correlation dimension as a discriminating value for the normal and abnormal fetuses.
RESULTS
After low pass filtering of fetal heart rate signal at the cutoff frequency, the numerical result of embedding dimension was 5.47+/-0.67 (normal), 4.43+/-0.60 (abnormal) (p<0.001), and correlation dimension was 2.51+/-0.43 (normal), 1.79+/-0.15 (abnormal).
CONCLUSION
We could determine the cutoff frequency from the different fluctuation level between high frequency noise and principal signal according the cutoff frequency function. The detection of fetal distress was done objectively by this method. We could discriminate between normal fetus and abnormal fetus by correlation dimension as indicator for the nonlinear complexity.

Keyword

Fetal heart rate; Measurement noise; Nonlinear time series analysis; Embedding dimension; Correlation dimension; Fetal distress

MeSH Terms

Female
Fetal Distress
Fetal Heart*
Fetus
Heart Rate, Fetal*
Humans
Hydrogen-Ion Concentration
Noise*
Pregnancy
Umbilical Arteries
Full Text Links
  • KJOG
Actions
Cited
CITED
export Copy
Close
Share
  • Twitter
  • Facebook
Similar articles
Copyright © 2024 by Korean Association of Medical Journal Editors. All rights reserved.     E-mail: koreamed@kamje.or.kr