Abstract—Noise removal in Magnetic Resonance (MR) images is important and essential for a wide variety of subsequent processing applications. Among the abundant denoising algorithms, the bilateral filter has been widely used in many image preprocessing procedures. However, it requires laborious tuning of parameters to obtain optimal filtering results, which is tedious and time-consuming. To address this problem, this paper is in an attempt to automate the bilateral filter based on an artificial neural network. Seven most significant attributes among 60 image attributes are used as the network input arguments. The BrainWeb image data with various scenarios of noise level, intensity non-uniformity, and slice thickness were adopted to evaluate this new system. Experimental results indicated that our automatic bilateral filter accurately predicted the denoising parameters and effectively removed the noise in MR images.