Quite often 3 anisotropic MRI images are acquired (high resolution along two axes, very low resolution along a third axis) to reduce time spent in the MRI scanner. However, these images are not well suited for 3D segmentation. See more information here: Combining volumes - what am I missing? . To reduce artifacts caused by low-resolution third axis you can follow the technique explained here: Segmenting Issue: I Can’t Remove “ridges” in 3-D model