Fast Region of Interest detection for fetal genital organs in B-mode ultrasound images

Genital organ detection of fetus in B-mode ultrasound images has a considerable significance. It is useful to know any malformations present in the genital organs and also to determine the sex of the fetus. In this paper we propose a Feature from Accelerated Segment Test (FAST) technique for approximate detection of fetal genitals in ultrasound images. FAST algorithm is capable of producing the corner points at a higher speed which falls on the fetal genital organs. A window of size 60×60 pixels being corner point as a center is considered as Region of Interest (ROI), where genital organ of fetus is anticipated. The efficiency of the algorithm is calculated as the ratio of number of images where corner points are placed on the fetus genital organ to the total number of images tested. FAST algorithm is robust to speckles present in the image, machine independent, fast and also computationally less intensive to implement in real time with an efficiency of 96.7%.