Development of Algorithm for Estimation of Gestational Age using each Variable of Fetal Heart Rate

Cha, Kyung Joon; Park, Young Sun; Ryu, Je Seon; Yoon, Sang Soon; Hwang, Jung Hae; Chung, Sung Ro; Moon, Hyung; Park, Moon Il

Abstract

OBJECTIVE
The objective of this study is to estimate gestational age by using FHR parameters after linear and nonlinear analysis of FHR data. Linear spectral decomposition has a limit to describe and analyze the nonlinear and complex physiological nature. Random processing and chaotic analysis assist to quantify some patterned energies, interaction of the physiological system in body system, which the constancy is preserved, and complexity of physical cardiovascular system. But there are no great development of new device which notify fetal age.
METHODS
From March 1995 to December 2000, 2,548 cases who received NST during antenatal visit were included in this study. Among these cases, 1000 cases were sampled randomly. We divided this cases into a study and a control group, 500 cases for model building set and the other 500 cases for validation set. Each FHR parameters such as baseline FHR, variability (AMP, MMR), acceleration & deceleration (15 bpm-15 seconds), loss of signal, the number of fetal movement were analysed by using our own computerized HYFM-I, II software system. The linear and non-linear analysis of FHR were done after extracting approximate entropy (ApEn) value. Finally, we performed regression analysis and extract an equation for estimation of gestational weeks using bootstrap method.
RESULTS
We extract a following equation using above study method: Log(GA)=5.870-0.051(Sloss*)-0.065(Mean FHR*)+0.049(Mean FHR*)2-0.058(FM*)+0.048(AMP*)2+0.121(A1515*)-0.031(A1515*)2+0.036(ApEn*)2.
CONCLUSIONS
Accurate dating of pregnancy, namely gestational week is very important for antenatal diagnosis in the field of perinatal medicine. This study will propose scientific research results which are useful for studying normal and morbid generating physiological fetal condition. We extract an appropriate equation for estimation of gestational weeks only using FHR parameters. We hope to compare our result with other authors' results in the near future.