Segmentation and estimation of brain tumor volume in magnetic resonance images based on T2-weighted using hidden markov random field algorithm

A brain tumor is an abnormal growth of tissue in the brain. the segmentation of brain tumors,which has been manually achieved from magnetic resonance images (mri) is a decisive and time-consuming task. treatment,diagnosis,signs and symptoms of the brain tumors mainly depend on the tumor size,position,and growth pattern. the accuracy and timeliness of detecting a brain tumor are vital factors to achieve the success in diagnosis and treatment of brain tumor. therefore,segmentation and estimation of volume of brain tumor have been deemed a challenge mission in medical image processing. this paper aims to present a new approach,to improve the segmentation of brain tumors form t2-weighted mri images using hidden markov random fields (hmrf) and threshold method. we calculate the volume of the tumor using a new approach based on 2d images measurements and voxel space. the accuracy of segmentation is computed by using the roc method. in order to validate the proposed approach a comparison is achieved with a manual method using mango software. this comparison reveals that the noise or impurities in measurement of tumor volume are less in the proposed approach than in mango software.