Extraction of Fetal Biometrics Using Class Separable Shape Sensitive Approach for Gestational Age Estimation

摘要

Extraction of fetal biometric parameters for accurate estimation of gestational age using class separable shape sensitive segmentation scheme is presented in this paper. The class separablity is achieved by introducing a perturbation with an appropriate cost functional associated to the image. The algorithm has been used to segment the gestational sac, yolk sac, femur bone and the parietal bones for gestational age estimation. It has been found that, Using the proposed scheme, it is possible to estimate the gestational age of the fetus with a prediction accuracy of ±2 days which is low compared to other traditional methods.