One of the most common methods for diagnosing the brain is using magnetic resonance imaging (MRI). The segmentation of brain MRI images using a computer pre-processing system is an important step in the work of applications aimed at planning surgical operations, studying brain activity or detecting malignant tumors. The primary goal of such a system is to improve image quality. In this paper, the main methods and algorithms for preliminary processing of MRI images are considered. The analysis of the methods and methods of noise suppression in MRI images showed that to solve this problem, such filters as averaging, Gauss, Wiener, median and Haar wavelet are most often used. Image preprocessing methods allow you to convert images to improve their visual perception, as well as solve problems of changing the presentation of images to ensure their storage, transmission, visualization in electronic form and further analysis of the information embedded in them.