You can use a 3D Fourier transform and discard high frequency components and reconstruct the image. That would remove the small, irrelevant regions.

 The unwanted parts are from various portion like vein,artery,blood vessel which are spread around the main mass portion. Now when image enhancement is done, then these unwanted parts are also enhanced with the main mass portion, so these parts are to be extracted out from the enhanced mass image. The size ,shape distribution are must be different for different blood vessel which are unknown for different image.so how can I extract these parts from main image in matlab automatically? 

http://www.mathworks.com/matlabcentral/fileexchange/56365-generalized-framework-cell-segmentation

In the above link, please find the function zhenzhou_shape_filtering() and try it. 

I hope this will help you to solve your problem. 

a=bwareaopen(a,'specified the size'); ; size of the unwanted region

a=imfill(a,'hole');

This may help you for the removal of unwanted region

try to use any enhancement or morphological operations for your issues

In my opinion two ways we can do this :

1. Find out all the connected components and remove the connected component with

try to use segmentation or morphological operations

thank you to all for giving most valuable advice.

thank you very much

Hi all I have a similar situation where this command is not allowing me to move forward. its a bit challenging and I am still trying to find a solution that would solve the conundrum! can anyone assist me please? thanks in advance!

I have tried the conn, and the P in this regard still produces an error!

this my code;

R3bw = imbinarize(R3); R3bw1 = bwareaopen(R3bw,50,);