Brain Tumor Detection and Segmentation Using Conditional Random Field

Medical image processing is a highly challenging field. Medical image techniques are used to mage the inner portions of the human body for medical diagnosis. MR Images are widely used in the diagnosis of brain tumor. In this paper, we present an automated method to detect and segment the brain tumor regions. The proposed method consists of three main steps, initial segmentation, modeling of energy function and optimize the energy function. To make our segmentation more reliable we use information present in the T1 and FLAIR MRI images. We use conditional random field(CRF) based frame work to combined the information present in T1 and FLAIR in probabilistic domain. Main advantages of CRF based frame work is we can mode complex shapes easily and we incorporate the observation of energy function.