A two phase segmentation algorithm for MRI brain tumor extraction

Abstract

Abnormal growth of cells in proliferated and uncontrolled manner leads to brain tumor. Accurate segmentation of brain tumor plays a vital role for analyzes, diagnosis and planning for treatment. An efficient method for brain tumor segmentation from T1 weighted MRI images is performed in two phases based on histogram thresholding and region growing techniques. Though simple and quick, thresholding alone cannot give accurate result so the image initially segmented using histogram thresholding is further analyzed and seed for region growing is automatically extracted for accurate tumor extraction in the subsequent phase for accurate tumor extraction.