Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{193692,
        author = {S. Aarthi and Dr. S. Chitra},
        title = {ICA-Based Preprocessing for Morphology-Preserving MRI Denoising},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {973-978},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=193692},
        abstract = {Reliable brain-tumor classification depends on high-quality MRI images. Excessive noise and artifacts can obscure tumor boundaries, hindering model performance. This study uses Independent Component Analysis (ICA) to preprocess brain-tumor MR images, which includes 253 grayscale, 64×64pixel slices (155 with tumors, 98 normal). ICA separates images into independent components; low-energy patterns are removed, and the rest are combined to produce denoised images with clearer contrast and preserved structural details. Quantitative results show improvements with an average PSNR of about 37Db and low reconstruction errors, demonstrating effective noise suppression without losing lesion morphology. This preprocessing produces consistent, high-fidelity inputs for hybrid CNN and graph-based classification, ensuring robust, noise-resistant, and interpretable brain-tumor analysis.},
        keywords = {Independent Component Analysis (ICA); MRI Denoising; Brain Tumor Classification; Preprocessing; Noise Suppression; Peak Signal-to-Noise Ratio (PSNR); Morphology Preservation; Feature Enhancement; Deep Learning; Medical Image Analysis},
        month = {March},
        }