دانلود Detection of Brain Tumor in 3D MRI Images using Local Binary Patterns and Histogram Orientation Gradient

Abstract

Brain tumor pathology is one of the most common mortality issues considered as an essential  priority for health care societies. Accurate diagnosis of the type of disorder is crucial to make a  plan  for  remedy  that  can  minimize  the  deadly  results.  The  main  purpose  of  segmentation and detection is to make distinction between different regions of the brain. Besides accuracy, these techniques should be implemented quickly.  In  this  paper  an  automatic  method  for  brain  tumor  detection  in  3D  images  has  been proposed. In the first step, the bias field correction and histogram matching are used for preprocessing  of  the  images.  In  the  next  step,  the  region  of  interest  is  identified  and  separated from  the  background  of  the  Flair  image.  Local  binary  pattern  in  three  orthogonal  planes (LBP-TOP)  and  histogram  of  orientation  gradients  (HOG-TOP)  are  used  as  the  learning features. Since 3D images are used in this research we use the idea of in local binary pattern in three orthogonal planes in order to extend histogram orientation gradients for 3D images. The random forest is  then used to segment tumorous regions. We evaluate the performance of  our  algorithm  on  glioma  images  from  BRATS  2013.  Our  experimental  results  and analyses  indicate  that  our  proposed  framework  is  superior  in  detecting  brain  tumors  in comparison with other techniques.