Menu 
VPA
Computer Vision And Pattern Analysis Laboratory Home Page  Home
People  People
Publications  Publications
Publications  Databases
Contact Information  Contact
Research
Supported Research Projects  Supported Research Projects
Research Activites  Research Activites
Research Groups
SPIS - Signal Processing and Information Systems Lab.SPIS - Signal Processing and Information Systems Lab.
Medical Vision and Analysis Group  Medical Research Activities
Biometrics Research Group  Biometrics Research Group
SPIS - Signal Processing and Information Systems Lab.MISAM - Machine Intelligence for Speech Audio and Multimedia.
Knowledge Base
  Paper Library
  VPA Lab Inventory
  Databases in VPALAB
Multimodal Brain Tumor Segmentation Using The “Tumor-cut” Method on The BraTS Dataset
Authors: Andac Hamamci, Gozde Unal
Published in: MICCAI 2012-Multimodal Brain Tumor Segmentation Challenge
Publication year: 2012
Abstract: In this paper, the tumor segmentation method used is described and the experimental results obtained are reported for the “BraTS 2012 - Multimodal Brain Tumor Segmentation Challenge” of MICCAI’12. “Tumor-cut” method, presented in [1] is adapted to multi-modal data to include edema segmentation. The method is semi-automatic, requiring the user to draw the maximum diameter of the tumor, which takes about a minute user-interaction time per case. The typical run-time for each case is around 10-20 minutes depending on the size of the tumor. Overall Dice overlap with the expert segmentation is 0.36 ± 0.25 for the edema/infiltration and 0.69 ± 0.20 for the tumor region.
  download full paper
Download

Home Back