MR Images contain a significant amount of noise which is complex in nature and is inconsistent throughout the image. Hence it becomes important to study a detailed relation between different statistical parameters of the image. In this paper, a ratio of the mean of the corrupted pixel intensities and the brightest pixel is found to be constant for a wide range of input noise. Using this finding, a simple equation for SNR is derived and is compared with the standard formula of SNR calculation. The absolute difference was studied and observations are drawn. The proposed algorithm was implemented on synthetic images from BrainWeb data base using MATLAB 2015a. © 2019 IEEE.