Brain Tumour Detection Using Self Organising Maps | Original Article

ABSTRACT:

Brain tumour is one of the serious diseases so it is necessary to have accurate detection at an early stage. Generally by using CT-scan and MRI techniques visual examination is done by doctors for detection of parts of brain having tumour. In this paper self organizing map algorithm is used to carry out brain tumour detection from MRI image. In first stage, image pre-processing is done to remove file artifacts and unwanted skull part then median filter is used for noise reduction. In second stage, segmentation is performed on preprocessed image by self organizing maps method which partitions the image into different regions. SOM is an unsupervised algorithm having high diversity of data. This has been constructed mainly for identifying tissues including White Matter (WM), Grey Matter (GM), and Cerebrospinal Fluid (CSF). At the end of the process tumour is extracted from the MR image and its exact position and the shape is determined.