Noise Reduction in Brain Magnetic Resonance Imaging Using a Convolutional Autoencoder
Keywords:
Convolutional Autoencoder, MRI Denoising, Noise Reduction, Deep-Learning, Image EnhancementAbstract
In clinical practice, precise and high-quality brain Magnetic Resonance Imaging (MRI) is pivotal for diagnosing and formulating effective treatment strategies. The research objective is to assess the viability of employing a Convolutional Autoencoders (CAE) for the mitigating noise in brain MRI images. The focus is brain MRI images and the various types of noise (Salt and Pepper, Speckle, and Gaussian noise) that typically corrupt images and may lead to inaccuracies in diagnosis. The research also applies methods to artificially generate these noise types to represent real-world scenarios. Specifically, the dataset of brain MRI images is collected, pre-processed, and artificially exposed to various noise types to simulate the real-world conditions after the CAE model is used to reconstruct the corrupted images. The CAE is assessed for its high efficiency and efficacy using Mean Squared Error (MSE) and Peak Signal-to-Noise Ratio (PSNR). The results indicate that the CAE is very effective in removing noise, particularly Salt and Pepper noise. The model achieves a PSNR of 27.0687 dB and an MSE of 0.00216246 at the lowest noise level. The model also demonstrates stability under varying levels of Speckle noise. Although performance degrades as noise increases, the model continues to demonstrate potential for further refinement. The research furthers the CAE’s analytical potential by assessing its denoising capabilities across various noise types and levels. The research adds value by outlining recommendations to the medical imaging community while identifying the need for future research on different classifications of noise and advanced regularization methods.
References
[1] W. Y. Leong, “AI-generated brain scans for synthetic healthcare data,” in 2025 21st IEEE International Colloquium on Signal Processing & Its Applications (CSPA). Pulau Pinang, Malaysia: IEEE, Feb. 7–8, 2025, pp. 140–143.
[2] R. K. Manoj, A. N. S., and M. Batumalay, “Automated brain tumor analysis with multimodal fusion and augmented intelligence,” Journal of Applied Data Sciences, vol. 6, no. 2, pp. 1277–1290, 2025.
[3] J. Shedbalkar, K. Prabhushetty, and A. Inchalc, “A comparative analysis of filters for noise reduction and smoothening of brain MRI images,” in 2021 6th International Conference for Convergence in Technology (I2CT). Maharashtra, India: IEEE, April 2–4, 2021, pp. 1–6.
[4] S. Li, J. Zhou, D. Liang, and Q. Liu, “MRI denoising using progressively distribution-based neural network,” Magnetic Resonance Imaging, vol. 71, pp. 55–68, 2020.
[5] W. El-Shafai, S. Abd El-Nabi, E.-S. M. El-Rabaie, A. M. Ali, N. F. Soliman, A. D. Algarni, and F. E. Abd El-Samie, “Efficient deep-learningbased autoencoder denoising approach for medical image diagnosis,” Computers, Materials, & Continua, vol. 70, no. 3, pp. 6107–6125, 2022.
[6] K. Srinivasan and N. Muthu, “A comparative study and analysis of contrast enhancement algorithms for MRI brain image sequences,” in 2018 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT). Bengaluru, India: IEEE, July 10–12, 2018, pp. 1–7.
[7] K. Kagoiya and E. Mwangi, “A hybrid and adaptive non-local means wavelet based MRI denoising method with bilateral filter enhancement,” International Journal of Computer Applications, vol. 166, no. 10, pp. 1–7, 2017.
[8] Q. Zhang, C. Liu, and G. He, “An improved wiener filter based on adaptive SNR MRI image denoising algorithm,” in 2022 International Conference on Computing, Communication, Perception and Quantum Technology (CCPQT). Xiamen, China: IEEE, Aug. 5–7, 2022, pp. 164–168.
[9] Z. Li, Z. Li, H. Li, Q. Fan, K. L. Miller, W. Wu, A. S. Chaudhari, and Q. Tian, “Enhance the image: Super resolution using artificial intelligence in MRI,” 2024. [Online]. Available: http://arxiv.org/abs/2406.13625
[10] A. G. Cicimen, H. F. J. Tregidgo, M. Figini, E. Messaritaki, C. B. McNabb, M. Palombo, C. J. Evans, M. Cercignani, D. K. Jones, and D. C. Alexander, “Image quality transfer of diffusion MRI guided by high-resolution structural MRI,” in International Workshop on Computational Diffusion MRI. Marrakesh, Morocco: Springer, Oct. 6, 2024, pp. 106–118.
[11] T. Vyas, R. Yadav, C. Solanki, R. Darji, S. Desai, and S. Tanwar, “Deep learning-based scheme to diagnose Parkinson’s disease,” Expert Systems, vol. 39, no. 3, 2022.
[12] F. Bertacchini, R. Rizzo, E. Bilotta, P. Pantano, A. Luca, A. Mazzuca, A. Lopez, and Alzheimer’s Disease Neuroimaging Initiative, “Mid-sagittal plane detection for advanced physiological measurements in brain scans,” Physiological Measurement, vol. 40, no. 11, 2019.
[13] W. Krois, F. Palmisani, P. Gr¨opel, P. Feil, M. L. Metzelder, J. M. Patsch, and C. A. Reck-Burneo, “Assessment of sacral ratio in patients with anorectal malformations: Can magnetic resonance imaging replace conventional radiograph?” Journal of Pediatric Surgery, vol. 56, no. 11, pp. 1993–1997, 2021.
[14] J. Kugelman, D. Alonso-Caneiro, S. A. Read, S. J. Vincent, F. K. Chen, and M. J. Collins, “Effect of altered OCT image quality on deep learning boundary segmentation,” IEEE Access, vol. 8, pp. 43 537–43 553, 2020.
[15] V. Kamath, A. Renuka, V. G. Kini, and S. Prabhu, “Exploratory data preparation and model training process for Raspberry Pi-based object detection model deployments,” IEEE Access, vol. 12, pp. 45 423–45 441, 2024.
[16] M. B¨ohland, R. Bruch, S. B¨auerle, L. Rettenberger, and M. Reischl, “Improving generative adversarial networks for patch-based unpaired image-to-image translation,” IEEE Access, vol. 11, pp. 127 895–127 906, 2023.
[17] I. G. S. M. Diyasa, V. I. Sunarko, E. Y. Puspaningrum, V. Asy’ari, and M. Z. Ibrahim, “Op-timization of multi-section and partially augmented Magnetic Resonance Imaging (MRI) images for brain tumor classification using ResNet-50,” CommIT (Communication and Information Technology) Journal, vol. 19, no. 1, pp. 115–128, 2025.
[18] E. Delamare, X. Fu, Z. Huang, and J. Kim, “Panoramic imaging errors in machine learning model development: A systematic review,” Dentomaxillofacial Radiology, vol. 53, no. 3, pp. 165–172, 2024.
[19] I. G. S. M. Diyasa, B. Damanik, and M. R. Mahendra A., “Classification of pneumonia using a deep learning convolutional neural network,” in 2023 IEEE 9th Information Technology International Seminar (ITIS). Batu Malang, Indonesia: IEEE, Oct.18–20, 2023, pp. 1–5.
[20] H. Alsaif, R. Guesmi, B. M. Alshammari, T. Hamrouni, T. Guesmi, A. Alzamil, and L. Belguesmi, “A novel data augmentation-based brain tumor detection using convolutional neural network,” Applied Sciences, vol. 12, no. 8, pp. 1–16, 2022.
[21] A. S. Chauhan, J. Singh, S. Kumar, N. Saxena, M. Gupta, and P. Verma, “Design and assessment of improved convolutional neural network based brain tumor segmentation and classification system,” Journal of Integrated Science and Technology, vol. 12, no. 4, pp. 793–793, 2024.
[22] I. Kandel, M. Castelli, and L. Manzoni, “Brightness as an augmentation technique for image classification,” Emerging Science Journal, vol. 6, no. 4, pp. 881–892, 2022.
[23] E. H. Fuadi, A. R. Ruslim, P. W. K. Wardhana, and N. Yudistira, “Gated self-supervised learning for improving supervised learning,” in 2024 IEEE Conference on Artificial Intelligence (CAI). Singapore: IEEE, June 25–27, 2024, pp. 611–615.
[24] A. Biswas and M. S. Islam, “A hybrid deep CNNSVM approach for brain tumor classification,” Journal of Information Systems Engineering & Business Intelligence, vol. 9, no. 1, pp. 1–15, 2023.
[25] M. P. J. Van Der Loo and E. De Jonge, “Data validation,” in Wiley StatsRef: Statistics reference online. Wiley, 2020.
[26] I. Nagarajan and G. G. Lakshmi Priya, “Removal of noise in MRI images using a block differencebased filtering approach,” International Journal of Imaging Systems and Technology, vol. 30, no. 1, pp. 203–215, 2020.
[27] A. Krull, T. O. Buchholz, and F. Jug, “Noise2Void-learning denoising from single noisy images,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA: Computer Vision Foundation, June 16–20, 2019, pp. 2129–2137.
[28] W. Wu, M. Chen, Y. Xiang, Y. Zhang, and Y. Yang, “Recent progress in image denoising: A training strategy perspective,” IET Image Processing, vol. 17, no. 6, pp. 1627–1657, 2023.
[29] N. Rajeswaran, T. S. Lawrence, R. P. Ramkumar, and N. Thangadurai, “An efficient technique to remove Gaussian noise and improve the quality of magnetic resonance image,” International Journal of Innovative Technology and Exploring Engineering (IJITEE), vol. 8, no. 10, pp. 2375–2377, 2019.
[30] B. Perumal, R. S. Devi, and M. P. Rajasekaran, “Extermination methods of image noises: A review,” 3C Tecnolog´ıa: Glosas de Innovaci´on Aplicadas a la Pyme, no. Special Issue, pp. 243–259, 2021.
[31] B. Nisha and M. V. Jose, “DTMF: Decisionbased Trimmed Multimode approach Filter for denoising MRI images,” Soft Computing, vol. 28, no. 7, pp. 6327–6342, 2024.
[32] X. Wang, Y. Hua, E. Kodirov, D. A. Clifton, and N. M. Robertson, “IMAE for noise-robust learning: Mean absolute error does not treat examples equally and gradient magnitude’s variance matters,” 2019. [Online]. Available: http://arxiv.org/abs/1903.12141
[33] V. Karpukhin, O. Levy, J. Eisenstein, and M. Ghazvininejad, “Training on synthetic noise improves robustness to natural noise in machine translation,” in Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). Hong Kong, China: Association for Computational Linguistics, 2019, pp. 42–47.
[34] O. Grigas, R. Damaˇseviˇcius, and R. Maskeli¯unas, “Positive effect of super-resolved structural magnetic resonance imaging for mild cognitive impairment detection,” Brain Sciences, vol. 14, no. 4, pp. 1–26, 2024.
[35] R. Halim and G. Putra Kusuma, “Enhancing image clarity with the combined use of REDNet and attention channel module,” International Journal of Computing and Digital Systems, vol. 16, no. 1, pp. 213–223, 2024.
[36] L. Kong, Z. Li, Y. Liu, J. Zhang, M. Chen, Q. Zhou, X. Qi, X. W. Deng, and Y. Peng, “A generalized deep learning method for synthetic ct generation,” International Journal of Radiation Oncology, Biology, Physics, vol. 117, no. 2, 2023.
[37] P. Batarius, A. A. Sinlae, and E. F. Fahik, “Anal-ysis of the quality of natural dyes in weaving exposed to sunlight using MSE and PSNR parameters,” Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 6, no. 5, pp. 797–802, 2022.
[38] Y. Y. Al-Aboosi, R. S. Issa, and A. K. Jassim, “Image denosing in underwater acoustic noise using discrete wavelet transform with different noise level estimation,” TELKOMNIKA (Telecommunication Computing Electronics and Control), vol. 18, no. 3, pp. 1439–1446, 2020.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 I Gede Susrama Mas Diyasa, Pangestu Sandya Etniko Siagian, Eva Yulia Puspaningrum, Wan Suryani Wan Awang, Sayyidah Humairah, Deshinta Arrova Dewi
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
a. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License - Share Alike that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
b. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
c. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
Â
USER RIGHTS
All articles published Open Access will be immediately and permanently free for everyone to read and download. We are continuously working with our author communities to select the best choice of license options, currently being defined for this journal as follows: Creative Commons Attribution-Share Alike (CC BY-SA)
