A Novel Self-Supervised Swin Transformer with Wavelet ‎Feature Extraction for Early Alzheimer’s Disease Recognition

  • Authors

    • Jupalli Pushpakumari Department of Computer Science and Engineering (AIML & IOT), Vallurupalli Nageswara Rao Vignana Jyothi Institute of ‎ Engineering and Technology, ‎Hyderabad, Telangana, India
    • Janapareddy Uttara Alekhya Department of Advanced Computer Science and Engineering, Vignan’s Institute of Information Technology (A), Visakhapatnam, ‎ Andhra Pradesh, India
    • Jakkapu Nagalakshmidevi Department of Computer Science and Engineering, Aditya University, Surampalem, ‎Andhra Pradesh, India
    • Mangu Maanasa Department of Computer Science and Engineering, GMR Institute of Technology(A), Rajam, Andhra Pradesh, India
    • Vaddadi Vasudha Rani Department of Information Technology, GMR Institute of Technology(A), Rajam, Andhra Pradesh, India
    • M. Somasundara Rao Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, ‎Vaddeswaram, Guntur District, Andhra Pradesh, India
    • Chandra Sekhar Musinana Department of Information Technology, MVGR College of Engineering, Vizianagaram, Andhra Pradesh, India
    https://doi.org/10.14419/syxzbg95

    Received date: May 31, 2025

    Accepted date: June 25, 2025

    Published date: June 30, 2025

  • Alzheimer's Disease; MRI; Skull Stripping; Wavelet Transform; Swin Transformer; Self-Supervised Learning; Deep Learning; Transfer Learning.

  • Abstract

    Alzheimer's disease is a structural and functional brain alterations that induces memory loss and thinking problems, which leads to worsen ‎over time. It’s crucial to identify the problem early so that treatment can be provided on time and care can be more effective. In this research, ‎we acquaint’s a new deep learning technique that uses MRI brain scans to find whether the person is suffering with Alzheimer’s disease or ‎not. Our methodology integrates self-supervised learning where the model trains the scanned images of MRI without labelled data, ad-‎vanced feature extraction techniques, and a modern vision transformer model. Initially, we used the FSL-BET tool to extracts only the relat-‎ed MRI images of the brains by using FSL (FMRIB Software Library). Then, we use a multi-level Wavelet Transform to acquire major ‎features from the images and mainly focusing on texture and frequency data that helps to observe disease-related changes in the brain. After ‎that, we trained a Swin Transformer model using self-supervised learning so it could acquire knowledge from the data without human label-‎ling. Now, we regulate the model to categorize the MRI scans into four classes: Mild Demented, Moderate Demented, Non-Demented, and ‎Very Mild Demented. Finally, this research showed that this model got a highest accuracy of 97.8%. It performed well when compared to ‎other popular deep learning models like VGG19 and ResNet, especially when only a small amount of labelled data was available. The pri-‎mary benefit of wavelet features with self-supervised learning is the enhancement of the model's performance. The research presents a ro-‎bust and scalable method for early detection of Alzheimer’s disease via brain MRI scans‎.

  • References

    1. S. Pallawi and D. K. Singh, “Detection of Alzheimer’s Disease Stages Using Pre-Trained Deep Learning Approaches,” 2023 IEEE 5th International Conference on Cybernetics, Cognition and Machine Learning Applications (ICCCMLA), pp. 252–256, Oct. 2023, https://doi.org/10.1109/ICCCMLA58983.2023.10346730.
    2. S. Liu et al., “Generalizable deep learning model for early Alzheimer’s disease detection from structural MRIs,” Scientific Reports, vol. 12, no. 1, Oct. 2022, https://doi.org/10.1038/s41598-022-20674-x.
    3. M. Mujahid, A. Rehman, T. Alam, F. S. Alamri, S. M. Fati, and T. Saba, “An efficient ensemble approach for Alzheimer’s disease detection using an adaptive synthetic technique and deep learning,” Diagnostics, vol. 13, no. 15, p. 2489, Jul. 2023, https://doi.org/10.3390/diagnostics13152489.
    4. Rathour, “Alzheimer Disease Classification Using Improved VGG 19 Architecture,” 2024 2nd International Conference on Intelligent Data Commu-nication Technologies and Internet of Things (IDCIoT), pp. 921–924, Jan. 2024, https://doi.org/10.1109/IDCIoT59759.2024.10467710.
    5. K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2016, https://doi.org/10.1109/CVPR.2016.90.
    6. W. A. Shehri, “Alzheimer’s disease diagnosis and classification using deep learning techniques,” PeerJ Computer Science, vol. 8, p. e1177, Dec. 2022, https://doi.org/10.7717/peerj-cs.1177.
    7. F. Yi et al., “XGBoost-SHAP-based interpretable diagnostic framework for alzheimer’s disease,” BMC Medical Informatics and Decision Making, vol. 23, no. 1, Jul. 2023, https://doi.org/10.1186/s12911-023-02238-9.
    8. Y. Huang and W. Li, “Resizer SWIN Transformer-Based Classification using SMRI for Alzheimer’s Disease,” Applied Sciences, vol. 13, no. 16, p. 9310, Aug. 2023, https://doi.org/10.3390/app13169310.
    9. J. Xin, A. Wang, R. Guo, W. Liu, and X. Tang, “CNN and swin-transformer based efficient model for Alzheimer’s disease diagnosis with sMRI,” Biomedical Signal Processing and Control, vol. 86, p. 105189, Jun. 2023, https://doi.org/10.1016/j.bspc.2023.105189.
    10. N. Garg, M. S. Choudhry, and R. M. Bodade, “Alzheimer’s disease detection through wavelet-based shifted elliptical local binary pattern,” Biomedi-cal Signal Processing and Control, vol. 100, p. 107067, Oct. 2024, https://doi.org/10.1016/j.bspc.2024.107067.
    11. P. A. Menon and Dr. R. Gunasundari, “Deep Feature Extraction and Classification of Alzheimer’s Disease: A Novel Fusion of Vision Transformer-DenseNet Approach with Visualization,” Journal of Internet Services and Information Security, vol. 14, no. 4, pp. 462–483, Nov. 2024, https://doi.org/10.58346/JISIS.2024.I4.029.
    12. S. Yang, J. M. S. Bornot, R. B. Fernandez, F. Deravi, K. Wong-Lin, and G. Prasad, “Integrated space–frequency–time domain feature extraction for MEG-based Alzheimer’s disease classification,” Brain Informatics, vol. 8, no. 1, Nov. 2021, https://doi.org/10.1186/s40708-021-00145-1.
    13. D. Nguyen et al., “Ensemble learning using traditional machine learning and deep neural network for diagnosis of Alzheimer’s disease,” IBRO Neu-roscience Reports, vol. 13, pp. 255–263, Sep. 2022, https://doi.org/10.1016/j.ibneur.2022.08.010.
    14. R. Khan et al., “A transfer learning approach for multiclass classification of Alzheimer’s disease using MRI images,” Frontiers in Neuroscience, vol. 16, Jan. 2023, https://doi.org/10.3389/fnins.2022.1050777.
    15. H. Wang, C. Lei, D. Zhao, L. Gao, and J. Gao, “DeepHipp: accurate segmentation of hippocampus using 3D dense-block based on attention mecha-nism,” BMC Medical Imaging, vol. 23, no. 1, Oct. 2023, https://doi.org/10.1186/s12880-023-01103-5.
    16. N. Raza, A. Naseer, M. Tamoor, and K. Zafar, “Alzheimer Disease Classification through Transfer Learning Approach,” Diagnostics, vol. 13, no. 4, p. 801, Feb. 2023, https://doi.org/10.3390/diagnostics13040801.
    17. Balaji, C. ., & Veni, S. . “Automatic Skull Stripping from MRI of Human Brain using Deep Learning Framework for the Diagnosis of Brain Related Diseases”, International Journal of Intelligent Systems and Applications in Engineering, vol. 11, no.4, Mar. 2023, pg.no. 439–445.
    18. M. S. Rao, S. K. UmaMaheswaran, N. C. Sattaru, K. H. Abdullah, U. K. Pandey, and L. Biban, “A critical understanding of integrated artificial intel-ligence techniques for the healthcare prediction system,” 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE), pp. 728–731, Apr. 2022, https://doi.org/10.1109/ICACITE53722.2022.9823678.
    19. S. Veluchamy, R. Bhuvaneswari, K. Ashwini, S. Alshathri, and W. El-Shafai, “HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system,” Engineering Science and Technology an International Journal, vol. 68, p. 102058, May 2025, https://doi.org/10.1016/j.jestch.2025.102058.
    20. I. Georgiana et al., “Transfer Learning for Alzheimer’s Disease Diagnosis from MRI Slices: A Comparative Study of Deep Learning Models”, 27th international conference on knowledge – Based and Intelligent Information & Engineering Systems, 2023, pg.no. 2614 – 2623. https://doi.org/10.1016/j.procs.2023.10.253.
    21. P. Zhang, S. Lin, J. Qiao, and Y. Tu, “Diagnosis of Alzheimer’s Disease with Ensemble Learning Classifier and 3D Convolutional Neural Network,” Sensors, vol. 21, no. 22, p. 7634, Nov. 2021, https://doi.org/10.3390/s21227634.
    22. P. K. Sekharamantry, M. S. Rao, Y. Srinivas, and A. Uriti, “PSR-LeafNet: a deep learning framework for identifying medicinal plant leaves using sup-port vector machines,” Big Data and Cognitive Computing, vol. 8, no. 12, p. 176, Dec. 2024, https://doi.org/10.3390/bdcc8120176.

  • Downloads

  • How to Cite

    Pushpakumari , J. ., Alekhya , J. U. ., Nagalakshmidevi , J. ., Maanasa , M. ., Rani , V. V. ., Rao, M. S. ., & Musinana, C. S. . (2025). A Novel Self-Supervised Swin Transformer with Wavelet ‎Feature Extraction for Early Alzheimer’s Disease Recognition. International Journal of Basic and Applied Sciences, 14(2), 542-550. https://doi.org/10.14419/syxzbg95