A Foundational Approach to Biomedical Image Enhancement for Brain Tumor Detection

The treatment of biomedical images through preprocessing and enhancement techniques draws the attention of everyone in the field of medical imaging due to the fact that such treatments will contribute to the improvement of the diagnostic accuracy. The present study investigates preprocessing and enhancement techniques applied to MRI brain tumor imaging. The tumor regions, consequently, become more visible and clearer. The methods used are Gaussian filtering, adaptive histogram equalization, and wavelet-based enhancement that remove the noise and make the image brighter at the same time. The power of the combination of these methods in yielding great images was astounding; the image quality measures used registered very significant changes—Peak Signal-to-Noise Ratio (PSNR) increased from 28,6 dB to 36,8 dB, Structural Similarity Index (SSIM) from 0,82 to 0,94, and Mean Squared Error (MSE) decreased by approximately 40 %. The tumor boundary delineation got better as a result of such advancements leading to more accurate segmentation and classification. This research verified that good preprocessing and enhancement were prerequisites for the reliable analysis of brain MRI images, besides, it pointed out the future directions of diagnostic and machine learning applications in the area of biomedical image processing.