Kidney CT image segmentation using multi-feature EM algorithm, based on Gabor filters

Abstract

Image segmentation plays a crucial role in many medical imaging applications by automating or facilitating the delineation of anatomical structures and other regions of interest. Computed tomography (CT) are images with low contrast and with heavy noise. To handle these types of images for the purpose of kidney tumor delineation, we propose a new automatic segmentation method using a multi-feature EM algorithm, based on texture information. The approach consists of two steps: finding an effective and discriminative set of texture features using Gabor filters and the EM based image segmentation. The experimental results show that our proposed method works well for both: 2D and 3D CT images.