21 March 2014 Brain tumor locating in 3D MR volume using symmetry
Author Affiliations +
Abstract
This work deals with the automatic determination of a brain tumor location in 3D magnetic resonance volumes. The aim of this work is not the precise segmentation of the tumor and its parts but only the detection of its location. This work is the first step in the tumor segmentation process, an important topic in neuro-image processing. The algorithm expects 3D magnetic resonance volumes of brain containing a tumor. The detection is based on locating the area that breaks the left-right symmetry of the brain. This is done by multi-resolution comparing of corresponding regions in left and right hemisphere. The output of the computation is the probabilistic map of the tumor location. The created algorithm was tested on 80 volumes from publicly available BRATS databases containing 3D brain volumes afflicted by a brain tumor. These pathological structures had various sizes and shapes and were located in various parts of the brain. The locating performance of the algorithm was 85% for T1-weighted volumes, 91% for T1-weighted contrast enhanced volumes, 96% for FLAIR and T2-wieghted volumes and 95% for their combinations.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Pavel Dvorak, Pavel Dvorak, Karel Bartusek, Karel Bartusek, } "Brain tumor locating in 3D MR volume using symmetry", Proc. SPIE 9034, Medical Imaging 2014: Image Processing, 903432 (21 March 2014); doi: 10.1117/12.2042845; https://doi.org/10.1117/12.2042845
PROCEEDINGS
6 PAGES


SHARE
Back to Top