Efficient Medical Image Compression using a CNN and Discrete Wavelet Transform
DOI:
https://doi.org/10.66950/xt6hw312Keywords:
image compression, convolutional neural network (CNN), discrete wavelet transform (DWT), medical imageAbstract
This study presents a computationally efficient framework for medical image compression that integrates a low-complexity convolutional neural network (CNN) with the discrete wavelet transform (DWT). Unlike conventional CNN–DWT hybrid approaches that embed wavelet processing within large end-to-end architectures, the proposed method maps the input image to a compact three-channel latent representation that is encoded using the CDF 9/7 wavelet, as employed in JPEG 2000. This design avoids custom coding of high-dimensional feature tensors while facilitating compatibility with established image-coding workflows. The CNN reduces spatial redundancy while preserving essential image information, whereas the DWT exploits frequency-domain sparsity and multiresolution characteristics to improve compression efficiency. A two-stage training strategy is adopted: the encoder–decoder is first pretrained without the codec and subsequently fine-tuned with real non-differentiable quantization in the compression loop, enabling optimization against the actual coding process. Experiments on a publicly available low-grade glioma (LGG), magnetic resonance imaging (MRI) dataset demonstrate compression ratios of up to 40:1, with a peak signal-to-noise ratio (PSNR) of 38.6 dB and a multiscale structural similarity index measure (MS-SSIM) of 98.84%. The proposed CNN contains approximately 0.26 million trainable parameters and can be trained on a single GPU, demonstrating its potential for resource-constrained medical image storage and transmission applications.
References
(1) P. K. Sran, S. Gupta, and S. Singh, “Segmentation based image compression of brain magnetic resonance images using visual saliency,” Biomedical Signal Processing and Control, vol. 62, Art. N°. 102089, Sep. 2020, https://doi.org/10.1016/j.bspc.2020.102089 DOI: https://doi.org/10.1016/j.bspc.2020.102089
(2) Z. Li, A. Ramos, Z. Li, M. L. Osborn, W. Zaid, X. Li, Y. Li, J. Xu “Nearly-lossless-to-lossy medical image compression by the optimized JPEGXT and JPEG algorithms through the anatomical regions of interest,” Biomedical Signal Processing and Control, vol. 83, Art. Numb. 104711, May 2023, https://doi.org/10.1016/j.bspc.2023.104711 DOI: https://doi.org/10.1016/j.bspc.2023.104711
(3) X. Liu., L. Zhang, Z. Guo, T. Han, M. Ju, B. Xu, H. Liu “Medical Image Compression Based on Variational Autoencoder,” Mathematical Problems in Engineering, vol. 2022, Art. Numb. 7088137. Dec. 2022, https://doi.org/10.1155/2022/7088137 DOI: https://doi.org/10.1155/2022/7088137
(4) S. Aumann, S. Donner, J. Fischer, and F. Müller, “Optical Coherence Tomography (OCT): Principle and Technical Realization,” in: J. F. Bille, "High Resolution Imaging in Microscopy and Ophthalmology: New Frontiers in Biomedical Optics", Ed., Cham: Springer International Publishing, 2019, pp. 59–85. https://doi.org/10.1007/978-3-030-16638-0_3 DOI: https://doi.org/10.1007/978-3-030-16638-0_3
(5) E. Aldemir, G. Tohumoglu , and M. A. and Selver, “Binary medical image compression using the volumetric run-length approach,” The Imaging Science Journal, vol. 67, no. 3, Apr. 2019, pp. 123–135, https://doi.org/10.1080/13682199.2019.1565695 DOI: https://doi.org/10.1080/13682199.2019.1565695
(6) M. K. Abdmouleh, H. Amri, A. Khalfallah, and M. S. Bouhlel, “A Fast JPEG2000 Based Crypto-Compression Algorithm: Application to the Security for Transmission of Medical Images,” In: Balas, V., Jain, L., Balas, M. (eds) Soft Computing Applications. SOFA 2016. Advances in Intelligent Systems and Computing, vol 633. Springer, Cham pp. 164–175. https://doi.org/10.1007/978-3-319-62521-8_14 DOI: https://doi.org/10.1007/978-3-319-62521-8_14
(7) C. P. Devadoss and B. Sankaragomathi, “Near lossless medical image compression using block BWT–MTF and hybrid fractal compression techniques,” Cluster Comput, vol. 22, no. 5, pp. 12929–12937, Sep. 2019, https://doi.org/10.1007/s10586-018-1801-3 DOI: https://doi.org/10.1007/s10586-018-1801-3
(8) P. Sreenivasulu and S. Varadarajan, “An Efficient Lossless ROI Image Compression Using Wavelet-Based Modified Region Growing Algorithm,” Journal of Intelligent Systems, vol. 29, no. 1, Jan. 2020, pp. 1063–1078, https://doi.org/10.1515/jisys-2018-0180 DOI: https://doi.org/10.1515/jisys-2018-0180
(9) N. H. Hussein and M. A. Ali, “Medical Image Compression and Encryption Using Adaptive Arithmetic: Coding, Quantization Technique and RSA in DWT Domain,” Iraqi Journal of Science, Vol. 63, N°. 5 pp. 2279–2296, May 2022, https://doi.org/10.24996/ijs.2022.63.5.38 DOI: https://doi.org/10.24996/ijs.2022.63.5.38
(10) J. Ballé, V. Laparra, and E. P. Simoncelli, “End-to-end Optimized Image Compression,” in 5th International Conference on Learning Representations (ICLR), Toulon, France, Apr. 2017. arXiv:1611.01704, https://doi.org/10.48550/arXiv.1611.01704
(11) J. Ballé, D. Minnen, S. Singh, S. J. Hwang, and N. Johnston, “Variational image compression with a scale hyperprior,” in 6th International Conference on Learning Representations (ICLR), Vancouver, Canada, Apr. 2018. arXiv:1802.01436. https://doi.org/10.48550/arXiv.1802.01436.
(12) E. Agustsson, M. Tschannen, F. Mentzer, R. Timofte, and L. V. Gool, “Generative Adversarial Networks for Extreme Learned Image Compression,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 02 November 2019, Seoul, Korea (South), pp. 221–231. https://doi.org/10.1109/ICCV.2019.00031 DOI: https://doi.org/10.1109/ICCV.2019.00031
(13) S. Shukla and A. Srivastava, “Medical images Compression using convolutional neural network with LWT,” IJMCTR, vol. 6, no. 6, Art Numb. 265086, Jun. 2018, Accessed: Jan. 24, 2025. [Online]. Available: https://www.neliti.com/publications/265086/ DOI: https://doi.org/10.31873/IJMCTR.6.6.45
(14) A. Kar, S. Phani Krishna Karri, N. Ghosh, R. Sethuraman, and D. Sheet, “Fully Convolutional Model for Variable Bit Length and Lossy High Density Compression of Mammograms,” presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2018, pp. 2591–2594. Accessed: Jan. 24, 2025. [Online]. Available: https://openaccess.thecvf.com/content_cvpr_2018_workshops/w50/html/Kar_Fully_Convolutional_Model_CVPR_2018_paper.html
(15) D. Minnen, J. Ballé, and G. D. Toderici, “Joint Autoregressive and Hierarchical Priors for Learned Image Compression,” in in book "Advances in Neural Information Processing Systems", editors: S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi and R. Garnett. Vol. 31, Curran Associates, Inc., 2018. https://proceedings.neurips.cc/paper_files/paper/2018/file/53edebc543333dfbf7c5933af792c9c4-Paper.pdf
(16) C. Fu, B. Du, and L. Zhang, “Hybrid-context-based multi-prior entropy modeling for learned lossless image compression,” Pattern Recognition, vol. 155, art. Numb. 110632, Nov. 2024, https://doi.org/10.1016/j.patcog.2024.110632 DOI: https://doi.org/10.1016/j.patcog.2024.110632
(17) J. Lu, L. Zhang, X. Zhou, M. Li, W. Li, and S. Gu, “Learned Image Compression with Dictionary-Based Entropy Model,” in 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 10-17 June 2025, pp. 12850–12859. https://doi.org/10.1109/CVPR52734.2025.01199 DOI: https://doi.org/10.1109/CVPR52734.2025.01199
(18) S. Li, J. Lu, Y. Hu, L. S. Mattos, and Z. Li, “Towards scalable medical image compression using hybrid model analysis,” J Big Data, vol. 12, art Numb. 45, Feb. 2025, https://doi.org/10.1186/s40537-025-01073-1 DOI: https://doi.org/10.1186/s40537-025-01073-1
(19) P. V. Bindu and A. Jabeena, “ROI and Non-ROI Image Compression Using Optimal Zero Tree Wavelet and Enhanced Convolutional Neural Network for MRI Images,” SN COMPUTER. SCIENCE., vol. 5, art. numb. 38, Nov. 2023, https://doi.org/10.1007/s42979-023-02335-6 DOI: https://doi.org/10.1007/s42979-023-02335-6
(20) B. Perumal and M. P. Rajasekaran, “A hybrid discrete wavelet transform with neural network back propagation approach for efficient medical image compression,” in 2016 International Conference on Emerging Trends in Engineering, Technology and Science (ICETETS), Pudukkottai, India, 24-26 February 2016, pp. 1–5. https://doi.org/10.1109/ICETETS.2016.7603060 DOI: https://doi.org/10.1109/ICETETS.2016.7603060
(21) R. Mostafiz, M. S. Uddin, N.-A.- Alam, Md. Mahfuz Reza, and M. M. Rahman, “Covid-19 detection in chest X-ray through random forest classifier using a hybridization of deep CNN and DWT optimized features,” Journal of King Saud University - Computer and Information Sciences, vol. 34, no. 6, Part B, Jun. 2022, pp. 3226–3235, https://doi.org/10.1016/j.jksuci.2020.12.010 DOI: https://doi.org/10.1016/j.jksuci.2020.12.010
(22) M. Beladgham, Y. Habchi, M. Ben Aissa, and A. Taleb-Ahmed, “Medical video compression using bandelet based on lifting scheme and SPIHT coding: In search of high visual quality,” Informatics in Medicine Unlocked, vol. 17, p. 100244, 2019, https://doi.org/10.1016/j.imu.2019.100244 DOI: https://doi.org/10.1016/j.imu.2019.100244
(23) M. Buda, A. Saha, and M. A. Mazurowski, “Association of genomic subtypes of lower-grade gliomas with shape features automatically extracted by a deep learning algorithm,” Computers in Biology and Medicine, vol. 109, pp. 218–225, https://doi.org/10.1016/j.compbiomed.2019.05.002 DOI: https://doi.org/10.1016/j.compbiomed.2019.05.002
(24) S. S. Magar and B. Sridharan, “Hybrid image compression technique using oscillation concept & quasi fractal,” Health Technol., vol. 10,, an. 2020, pp. 313–320, J https://doi.org/10.1007/s12553-018-00282-4 DOI: https://doi.org/10.1007/s12553-018-00282-4
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Dalila Chikhaoui, Mohammed BELADGHAM , Mohamed Ben Aissa (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Authors retain copyright and grant the journal right of first publication.
