Abstract: In medical image processing, noise plays a role of reducing the details of the images and blurs the features which are important for the diagnosis of the disease. Brain images are fractal in nature and especially in brain MRI (Magnetic Resonance Imaging) images, fractional Brownian motion (fBm) noise affects the important features. To reduce the effect of fBm noise in brain MRI images by implementing wavelet based thresholding techniques namely visu shrink, SURE shrink and bayes shrink and to compare the performance of these techniques using various evaluation metrics. The fBm noise is greatly reduced in bayes shrink and it has high values of PSNR. This study provides, the implementation of wavelet domain thresholding techniques for denoising brain MRI images and provides a comparison of these shrink methods using PSNR (Peak Signal to Noise Ratio), MSE (Mean Square Error), absolute error, fractal dimension, IEF (Image Enhancement Factor), normalized cross correlation, structural content and much more.