3D Liver Tumor Segmentation in CT Images Using Improved Fuzzy C-Means and Graph Cuts

Algorithm 1

Kernelized FCM with spatial constraints (KFCMS).

input: the tumor seed point , the tumor VOI

output: the probabilistic models for tumor and liver

Set the number of clusters , weighting factor , , ,

neighborhood window of radius 3 voxels, the Gaussian parameter ,

maximum number of iterations , and maximum error

Set the intensity of the seed point, and the average intensity value of the liver

region as the initial cluster centers

  for iteration   do

Calculate the membership matrix according to (4)

Update the new cluster centers according to (6)

  , if ,

end for

Generate probabilistic model for each cluster derived from .