Fetal head segmentation based on Gaussian elliptical path optimize by flower pollination algorithm and cuckoo search

Number of maternal and infant mortality in Indonesia is high. This problem can be minimized by monitoring the fetal condition via ultrasound image. In addition, Indonesia have small number of obstetrics and gynecology compare to number of its population. Moreover, it is centralized in urban areas, so it is hard to monitor the condition of every babies in Indonesia. In order to resolve this problem, we have built fetal head monitoring system. Part of the system is to segment the fetal head in ultrasound image. In this paper, we examine nature optimization such as bat algorithm, cuckoo search, and flower pollination algorithm for optimizing Gaussian elliptical path for automatic fetal head segmentation. Experiment results shows that nature optimization Based Gaussian elliptical path (DoGEII-FPA and DoGEII-CS) has a minimum error compared to Gaussian elliptical path (DoGEll) which is optimized by Nelder-Mead. Interestingly, DoGEll-FPA and DoGEll-CS perform well from DoGEll-NM in different image.