K-means algorithm with level set for brain tumor segmentation

Samah Abdelaziz, Songfeng Lu


Brain is a complicated structure consisting of millions of millions cells so that, it’s difficult to identify any diseases without using any computerized technology. Magnetic resonance imaging (mri) is one of the main assessments of brain tumors. One of the most important steps on medical image processing is segmentation. Segmenting brain mri images, which provide accurate information for the diagnosis and therapy decisions of brain tumors. We proposed to segment brain tumor mri images into three parts (wm (white matter), gm (gray matter), and background). The first algorithm is for applying median filtering on brain mri image for removing the noise from the image for achieving accurate results. The second algorithm is for applying k-means algorithm for accuracy in time consuming and for clustering into regions and  the third algorithm indicate the detecting the boundary of the image with the use of level set. By comparison, our proposed method, its efficiency to segment perfectly more than other previous used algorithms especially on time consuming.


Image segmentation, Active contour model, Level set method, Brain tumor MRI;

Full Text:


DOI: http://doi.org/10.11591/ijeecs.v15.i2.pp991-1000


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics