Abstract: This paper proposes a new algorithm based on traditional fuzzy C-means algorithm regard
to the noise and uneven light in medical images. Fuzzy C-means clustering algorithm has been rapid
developed in image segmentation applications, as simple description, easy to implement, works well
for segmentation. But there are also other issues such as noise sensitive. Considering that the medical
images data must contain noise, a modified objective function J(u, v) has been proposed, adding a
punishment factor on the basis of introducing the pixel neighborhood information. The new algorithm
covers the shortage of traditional fuzzy C-means clustering algorithm, which makes the algorithm
clustering with noise more effectively. Experimental results show that the algorithm is effective and
practical.

Key words: fuzzy clustering, neighborhood pixels, punishment factor, medical image segmentation