This study focuses on enhancing the quality of MRI images of the cervix by applying and evaluating spatial domain image filtering techniques. MRI images were sourced from the CC-Tumor Heterogeneity database, and simulated thermal noise was introduced to mimic real-world conditions. Both linear and non-linear filters were applied to assess their effectiveness in noise reduction and detail preservation. The performance of the filters was quantitatively evaluated using PSNR (peak signal-to-noise ratio), SSIM (structural similarity index), and MSE (mean squared error) metrics. The anisotropic diffusion filter emerged as the most effective for noise removal, while the Laplacian filter excelled at edge enhancement. The results, presented through graphs and tables, highlight the most suitable filters for improving MRI images and provide recommendations for their application in clinical settings. Building on these findings, we explored some deep learning-based denoising techniques that address the limitations of traditional spatial domain filters.