پديدآورندگان :
Ebrahimian Homayoun Ebrahimian@iauardabil.ac.ir Department of Biomedica Engineering, Ardabil Branch, Islamic Azad University, Ardabil, Iran. , Hooshmandi Shahzad sh.hooshmandi@gmail.com Ardabil Branch, Islamic Azad University, Ardabil, Iran , Azhang Ayda Ayda372@yahoo.com Ardabil Branch, Islamic Azad University, Ardabil, Iran , Maleki Nazila Nazilamaleki1371@yahoo.com Ardabil Branch, Islamic Azad University, Ardabil, Iran
كليدواژه :
Digital image processing , Image segmentation , Medical images processing , MRI images , SVM
چكيده فارسي :
Automatically segmentation of brain tumor on MRI images is a very difficult and important task for the experts in brain surgery planning. If experts do the segmentation manually according to their own knowledge and experience, it will be time-consuming and may not reach any result. In this article, in order to minimize the time and reduce expert’s diagnosis errors, an automatic method for segmentation of brain tumor on MRI images based on SVM algorithm is presented. In this research, 30 T1-weighted MRI images are used from www.oasis-brains.org , because this type of images has higher contrast than T2-weighted and CT images. The proposed method segmented 26 tumor MRI images out of the target data and practically it achieved to 86.66% of segmentation among these 30 data.
