Ultrasound measurements in pregnancy involves measurement of fetal head and brain structures manually from 2D ultrasound volumes. This process required more time consuming. So here proposed automatic fetal head and brain (AFHB) system which measures anatomical structures from 3-D ultrasound volumes. The detection relies on Sequential Estimation techniques. The structure poses and embedded results obtained by this method can detect more than objects detected individually. The posterior distribution of the structure pose is approximated at each step by sequential Monte Carlo. This new model solves many challenges such as speckle noise, signal drop-out, shadows caused by bones, and appearance variations caused by the differences in the fetus gestational age. The detection process is done by Threshold based segmentation which is used to estimate sequentially. With this our proposed method the standard visualization plane are automatically detected and measured the value of an anatomy. The histogram equalization to image because some images have more brightness and some have low. This may cause imperfect segmentation result. In the high contrasted image, the low pitch cannot be determined and the segmentation cannot make perfect and in the low contrasted image. The statistical features are extracted from the segmented region. Finally neural network classifier is used for classify the fetal head to their growth and health stages.