Cluster Analysis for MRI Brain Tumor Segmentation


  • M. Rajmohan1, M. Abraham1 , D. Sahaya Lenin1 ,P. Pandiaraj1



Brain Tumor Segmentation, Discrete Wavelet Transform, Inverse DWT, K-Means Clustering


The abnormal tissues which are originated from the brain cells are known as brain tumor. It may be cancerous

tumor or non-cancerous tumor; these can cause pressure inside the skull to increase the tumors. This

damages the brain and become life-threatening. The main cause of brain tumor is still unknown. The early

diagnosis is required otherwise it increases the mortality rate. The segmentation of tumor part is essential

to identify the affected area in the brain. In this study, an effective method for Brain Tumor Segmentation

(BTS) is presented. The BTS system uses Discrete Wavelet Transform (DWT), K-Means Clustering (KMC)

algorithm and morphological operations for segmentation. Firstly, the input brain images are given to DWT

for decomposition and it produces lower and higher frequency sub-band coefficients. Then the Inverse

DWT (IDWT) is applied to reconstruct the image. The reconstructed image is given to KMC technique

for segmentation. Then the unwanted regions are removed by morphological operations to detect the brain

tumor in the given input image.

Author Biography

M. Rajmohan1, M. Abraham1 , D. Sahaya Lenin1 ,P. Pandiaraj1

1Assistant Professor, Hindustan Institute of Technology & Science, Chennai



How to Cite

M. Rajmohan1, M. Abraham1 , D. Sahaya Lenin1 ,P. Pandiaraj1. (2020). Cluster Analysis for MRI Brain Tumor Segmentation. Medico Legal Update, 20(3), 356-361.

Most read articles by the same author(s)