Comparative study of different segmentation techniques applied to medical imaging

In order to extract fetal body parts from ultrasound images, we tried to apply various image segmentation techniques such as Thresholding, Region Seed growing, K means, Fuzzy C means, parametric deformable model etc. onto the fetal ultrasound images. Results after performing segmentation are compared using Correlation, Mean Square Error, Peak Signal to Noise Ratio, Structural Similarity. It is observed that parametric active contour are providing better results compared to other techniques of segmentation that could be applied to ultrasound images.