Modified texture, intensity and orientation constraint based region growing segmentation of 2D MR brain tumor images

Abstract EN

Image segmentation is a process of dividing an image into different regions such that each region is nearly homogeneous.

Magnetic Resonance (MR) images always contain a significant amount of noise caused by operator performance, equipment and the environment which can lead to serious inaccuracies with segmentation.

Radiologists perform diagnosis manually at early stage.

It is a very challenging and difficult task for radiologists to correctly classify the abnormal regions in the brain tissue, because Magnetic Resonance Images (MRI) images are noisy images.

Because the tumors are inhomogeneous, un-sharp and faint, but show an intensity pattern that is different from the adjacent healthy tissue, a segmentation based on intensity, orientation and texture properties is proposed here.

With this approach the image segmentation problem can be formulated and solved in a principled way based on well-established mathematical theories.

The image clustering using texture also reduces the sensitivity to noise and results in enhanced image segmentation performance.

The ground truth of the tumor boundaries is manually extracted from publicly available sources.

Experimental results show that our method is robust and more accurate than other well known models.

The superiority of the proposed method is examined and demonstrated through a large number of experiments using MR images.