Automatic Tumor Classification of Brain MRI Images using DWT Features

V. Vani, M. Kalaiselvi Geetha


Brain tumor classification is an active research area in medical image processing and pattern recognition. Brain tumor is an abnormal
mass of tissue in which some cells grow and multiply uncontrollably, apparently unregulated by the mechanisms that control normal cells. The
growth of a tumor takes up space within the skull and interferes with normal brain activity. The detection of the tumor is very important in
earlier stages. Automating this process is a challenging task because of the high diversity in the appearance of tumor tissues among different
patients and in many cases similarity with the normal tissues. This paper depicts a novel framework for brain tumor classification based on
Discrete Wavelet Transform (DWT) features are extracted from the brain MRI images, which signify the important texture features of tumor
tissue. The experiments are carried out using BRATS dataset, considering three classes viz (Normal, Astrocytomas and Meaningiomas) and the
extracted features are modeled by Support Vector Machines (SVM), k-Nearest,Neighbor (k-NN) and Decision Tree (DT) for classifying tumor
types. In the experimental results, k-NN exhibit effectiveness of the proposed method with an overall accuracy rate of 85.45%, this outperforms
the SVM and DT classifiers.

Keywords—MRI, DWT, SVM, K-NN, DT, Brain Tumor, Tumor Types, BRATS.

Full Text:




  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science