BRAIN TUMOR DETECTION BASED ON ASYMMETRY AND K-MEANS CLUSTERING MRI IMAGE SEGMENTATION

Abstract

The brain is one of the largest and most complex organs of the human body. The brain can be a victim of numerous pathologies, including malignant tumours, strokes, infection, head injuries, and diseases. Brain tumour extraction and analysis are challenging tasks for medical image processing due to the complexity of images. Since the growth of tumours causes asymmetry in the affected parts of the brain, the proposed method calculates asymmetry based on the intensity difference between the left and right of a Mid-Sagittal Plane (MSP). One of the problems of this method appears when the brain object is rotated or tilted. A new method is proposed to solve this problem, by locating the MidSagittal Plane in T1-weighted MRI images, based on the low intensity of InterHemispheric Fissure (IF) region. In this paper, we have proposed segmentation of the brain MRI image using K-means clustering algorithm followed by a connected component label to determine the location and size of a tumour. The experimental result clearly shows the efficiency of the proposed method in comparison to the traditional systems in terms of computational cost and consumed time.

Keywords

• Bilateral symmetry
• Brain tumour
• Connected component labelling
• Kmeans clustering
• MRI