Analysis of denoising filters on MRI brain images

Abstract

The magnetic resonance imaging (MRI) modality is an effective tool in the diagnosis of the brain. These MR images are introduced with noise during acquisition which reduces the image quality and limits the accuracy in diagnosis. Elimination of noise in medical images is an important task in preprocessing and there exist different methods to eliminate noise in medical images. In this article, different denoising algorithms such as nonlocal means, principal component analysis, bilateral, and spatially adaptive nonlocal means (SANLM) filters are studied to eliminate noise in MR. Comparative analysis of these techniques have been with help of various metrics such as signal‐to‐noise ratio, peak signal‐to‐noise ratio (PSNR), mean squared error, root mean squared error, and structure similarity (SSIM). This comparative study shows that the SANLM denoising filter gives the best performance in terms of better PSNR and SSIM in visual interpretation. It also helps in clinical diagnosis of the brain.

An improved machine learning technique based on downsized KPCA for Alzheimer's disease classification

“…Some methods, after dimensionality reduction, use a small number of features although reducing features too much causes the accuracy rate to decrease . The most known algorithms for feature reduction are the principal component analysis (PCA), kernel PCA (KPCA), LDA and independent component analysis . The feature selection and reduction phase improves the performance of the model because it minimizes the consequences of the curse of dimensionality and the procedure of hastening learning.…”

MRI brain segmentation in combination of clustering methods with Markov random field

“…This article proposes a combination of GMM and HMRF due of the excellent capability of GMM method in MRI image segmentation and introduced a preprocessing step (MFCM) to reduce the computational burden of the system. There exist different preprocessing methods and the comparison among the filters has been provide with different performance metrics like Signal‐to‐Noise Ratio, Peak Signal‐to‐Noise Ratio, Mean Squared Error, Root Mean Squared Error, and Structure Similarity . The HMRF method is used in the post processing stage to reduce the search space to be smaller with less computational time.…”

Removal of noise in MRI images using a block difference‐based filtering approach

“…A threshold function is applied to prevent diffusion that happens across edges, and thus it preserves the edges in the image. Recently, it has been proposed that the spatial adaptive nonlocal mean (SANLM) 15 improves the quality of MRI images. Merely, it has certain limitations such as removing the small features of the image and adds the effect of the staircase to the filtered image.…”

“…Two weighing functions are designed for spatial and radiometric information, which are designed to modify a pixel value with an average of similar and nearby pixel values in a neighborhood. Recently, it has been proposed that the spatial adaptive nonlocal mean (SANLM) 15 improves the quality of MRI images. The noise level of an image can be eradicated spatially by this method.…”