Secure Hybrid Compression for Brain MRI Using DCT, DWT, a Convolutional Autoencoder, and Lightweight Encryption

Brain Magnetic Resonance Imaging (MRI) is critical for diagnosing neurological disorders, but the growing volume of MRI data poses challenges in storage, transmission, and security. Conventional compression techniques, including Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT), and formats such as JPEG, JPEG2000, and WebP, reduce file size but often degrade fine structural details, affecting clinical interpretation. This study presents a novel hybrid framework that integrates classical transforms, modern image formats, and a Convolutional Autoencoder (CAE) with a lightweight XOR-based transformation for secure handling of compressed MRI representations. The approach includes: baseline compression with DCT, DWT, and standard formats; CAE training with progressively increasing encoder filters (32–256) and a 1024-dimensional latent space; and XOR-based transformation of latent codes to provide a lightweight obfuscation layer during storage or transmission. Feature heatmaps extracted from the encoder validate the preservation of anatomical details.
The evaluation uses 253 MRI slices from a single publicly available dataset; therefore, further validation on larger and multi-institutional datasets is required and show that classical and standard methods achieve moderate compression (PSNR 23–27 dB) with artifacts, whereas the standalone CAE provides flexible compression (PSNR ~22.9 dB). The CAE with XOR-based transformation achieves high reconstruction fidelity, with a reported PSNR of 52.11 dB and SSIM of 0.994, shows low reconstruction error in the evaluated MRI slices, including regions containing visible tumor boundaries, and reduces storage by approximately 15:1. The framework simultaneously ensures high-quality compression and secure transmission, indicating potential applicability to medical image storage and transmission scenarios. Future work will explore deeper architectures, multimodal MRI datasets, advanced encryption methods, and real-time edge deployment.

Brain MRI, Image Compression, Convolutional Autoencoder, Secure Compression, Lightweight Encryption

1. Achhodawala, D. (2016). Analysis of DCT and DWT compression technique using JPEG image. Anveshana’s International Journal of Research in Engineering and Applied Sciences, 1(5), 1–6. [Google Scholar] [Crossref]

2. Akhtar, N., & Mian, A. (2022). Secure and efficient medical image compression using deep learning and encryption. IEEE Access, 10, 50612–50625. https://doi.org/10.1109/ACCESS.2022.3167289 [Google Scholar] [Crossref]

3. Awan, M. J., Iqbal, A., Khan, M. A., & Alghamdi, A. S. (2023). Blockchain-based privacy-preserving medical image transmission in telemedicine systems. Sensors, 23(5), 2498. https://doi.org/10.3390/s23052498 [Google Scholar] [Crossref]

4. Balle, J., Laparra, V., & Simoncelli, E. P. (2018). Variational image compression with a scale hyperprior. International Conference on Learning Representations (ICLR). https://arxiv.org/abs/1802.01436 [Google Scholar] [Crossref]

5. Chen, Y., Zhang, Q., Li, X., & Wang, T. (2022). Transformer-based medical image compression. Medical Image Analysis, 82, 102612. https://doi.org/10.1016/j.media.2022.102612 [Google Scholar] [Crossref]

6. Chowdhury, R. S., Paul, M., & Sharma, S. (2024). Improved DWT and IDWT architectures for image compression. Microprocessors and Microsystems, 104, 104990. https://doi.org/10.1016/j.micpro.2023.104990 [Google Scholar] [Crossref]

7. Huang, H., Li, Y., Zhang, Z., & Chen, Q. (2023). GAN-based high-fidelity medical image compression. Neural Computing and Applications, 35(12), 8491–8506. https://doi.org/10.1007/s00521-022-07627-1 [Google Scholar] [Crossref]

8. Islam, M. N., Rahman, M. A., & Hossain, M. S. (2021). An overview of homomorphic encryption for secure medical image analysis. Future Generation Computer Systems, 123, 1–14. https://doi.org/10.1016/j.future.2021.03.004 [Google Scholar] [Crossref]

9. Kaggle. (n.d.). Brain MRI images for brain tumor detection. https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection [Google Scholar] [Crossref]

10. Khan, M. A., Awan, M. J., & Raza, M. (2024). Lightweight deep learning frameworks for resource-constrained telemedicine. Computers in Biology and Medicine, 171, 107813. https://doi.org/10.1016/j.compbiomed.2024.107813 [Google Scholar] [Crossref]

11. Li, S., Zhang, H., & Wu, J. (2025). Towards scalable medical image compression using hybrid DWT and CNN. Journal of Big Data, 12(45). https://doi.org/10.1186/s40537-025-00999-x [Google Scholar] [Crossref]

12. Patel, B., & Achhodawala, D. (2016). Discrete cosine and wavelet transform techniques on JPEG picture compression techniques and functionalities. Anveshana’s International Journal of Research in Engineering and Applied Sciences, 1(6), 1–6. [Google Scholar] [Crossref]

13. Patel, J. R., & Achhodawala, D. (2017). Image compression run length encoding schema on RGB values. International Journal of Recent Scientific Research, 8(12), 22500–22504. [Google Scholar] [Crossref]

14. Rodrigues, M., Kormann, M., & Al-Dulaimi, M. (2016). Data protection and privacy issues concerning facial image processing in public spaces. Athens Journal of Technology & Engineering, 3(1), 39–52. https://doi.org/10.30958/ajte.3-1-3 [Google Scholar] [Crossref]

15. Raghu, R., Zhang, C., Kleinberg, J., & Bengio, S. (2019). Transfusion: Understanding transfer learning for medical imaging. Advances in Neural Information Processing Systems, 32. https://arxiv.org/abs/1902.07208 [Google Scholar] [Crossref]

16. Srivastava, S., Gupta, A., & Rathi, S. (2025). An efficient deep learning framework for detecting and classifying brain tumour from DWT compressed MRI images. Multimedia Tools and Applications. https://doi.org/10.1007/s11042-025-15287-9 [Google Scholar] [Crossref]

17. Toderici, G., Vincent, D., Johnston, N., et al. (2020). Full resolution image compression with recurrent neural networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(6), 1967–1981. https://doi.org/10.1109/TPAMI.2019.2910871 [Google Scholar] [Crossref]

18. Wang, C., Han, Y., & Wang, W. (2019). An end-to-end deep learning image compression framework based on semantic analysis. Applied Sciences, 9(17), 3580. https://doi.org/10.3390/app9173580 [Google Scholar] [Crossref]

19. Zaveri, S. H., & Achhodawala, D. (2016). Analysis of students’ enrollment in government and private schools in India using classification mining. Global Journal for Research Analysis, 5(4), 24–27. [Google Scholar] [Crossref]

• Explainable 3D Convolutional Neural Networks Spatiotemporal Learning for Human Handshake Interaction Recognition
• Comprehensive Empirical Benchmarking of Twelve Sorting Algorithms Across Comparison-Based, Non-Comparison-Based, and Hybrid Paradigms: A Multi-Dimensional Performance Model for Algorithm Selection at Practical Data Scales (n up to 100,000)
• Reinforcement Learning for Personalized Insulin Dosing: A Comparative Study of A2C, SAC and PPO on Real-World Clinical Data
• Faculty Milestone Manager and Appraisal System: A Web-Based Solution Part 1
• Improving Diabetes Prediction Through IQR Outlier Treatment an Evaluation of XGBoost And LightGBM Models