A robust and efficient approach for image denoising and brain region extraction to aid neurology system of patient

Abstract

Neurologist often tends to regard diseases of the nervous system as a difficult area. For patient presenting with symptoms of tumor should be diagnosed properly. Since treatment may not cure at the later stage, researchers must aim to produce maximal benefit to the patient with minimal burden, taking quality of survival into account as well as the duration. The computed tomography scan images are limited by the resolution of the imaging. In the field of Medical Resonance Image processing the image segmentation and denoising are very important and challenging problems in an image analysis. In this research paper the framelet transform for image denoising is implemented. Furthermore, the main purpose of segmentation in MRI images is to diagnose the problems in the normal brain anatomy and to find the location of tumor. This paper proposes a novel algorithm for segmentation of MRI images to extract the exact area of the brain as preprocessing steps for tumor location with image denoising. As a part of performance evaluation, 1000 images of patients are captured from different MRI centers under different conditions. Neuroradiological research consists of several brain extraction algorithms which are useful for several post- automatic image processing operations like segmentation, registration and compression. The result of proposed algorithm is validated by comparing proposed algorithm with the results of the existing segmentation and Denoising algorithms.