Segmentation and denoising of phase contrast MRI image of the aortic lumen via fractal and morphological processing

Magnetic Resonance Imaging (MRI) is a non-invasive widely used modality in medical diagnosis. One of disadvantages of this method is the low signal to noise ratio (SNR) and (or) contrast to noise ratio (CNR). That's why the image analysis tasks such as segmentation, reconstruction and registration are rather complicated. So the noise reduction in MR images is very important as a pre-processing task before going to image analysis. Furthermore, two dimensional (2D) imaging is incapable to provide a sufficient understanding of the complexity of the studied objects. So there is the necessity of having a 3D computer representation of many real objects. In this paper we are presenting the denoising and segmentation techniques developed for 3D MRI based on fractal and morphological approaches.