Design And Implementation of Efficient Brain Tumor Segmentation With Hybrid Fuzzy K-Means Clustering Technique

Abstract

This research is focused on detecting Brain tumors from MRI images using an interface of GUI in Matlab. Using  the GUI,  this  program  can  use  various combinations of  segmentation,  filters, and other  image processing algorithms  to achieve  the best  results. We  start  with  filtering  the  image  using  Prewitt  horizontal  edge-emphasizing  filter. The next step for detecting tumor is "watershed pixels". The  most  important  part  of  this  project  is  that  all  the  Matlab programs work with GUI Matlab guide. This allows us to use various combinations of filters, and other   image processing techniques to arrive at the best result that can help us detect brain tumors in their early stages. The development of biomedical image processing has gained attention from  the scientists,  there are some problems with biomedical image processing that have appeared, particularly with  MRI  imaging. The one general problem with  the biomedical  image  segmentation  (MRI  image)  method  for  image  segmentation  is  that  it  varies widely  depending  on  the  specific  application   For  example,   the  segmentation  program  that  is  used  with  MRI  imaging  has  different  requirements from segmentation of CT scan imaging. Furthermore, each image, even if it is from the same image application, such as MRI, has its own idiosyncrasies. These can be different from  other MRI  images, which when  read  by  the  same  segmentation  program will  give  a  different  result. Here in this research, we had proposed a novel MR brain image segmentation for detecting the tumor and to find the area of the tumor with improved accuracy and reduced computational time. This dissertation deals with the new hybrid clustering algorithm for reducing the computational time and binarization method to calculate the area in terms of ?mm?^2 based on the typography and digital imaging units. We compared the simulation results with the existing algorithms with the proposed shaft algorithm then after we found the area of tumor and calculated the CPU computational time. Finally, the proposed algorithm has performed far better than the existing algorithms with reduced computational time.