Deep Learning-Based Multi-Classification of Alzheimer’s Disease Stages Using Fine-Tuned VGG19 and MLP on MRI Scans
Abstract
Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder and a leading cause of dementia worldwide, making early and accurate diagnosis essential for effective clinical intervention. Although recent Deep Learning (DL) approaches have shown promising results for automated AD diagnosis from brain Magnetic Resonance Imaging (MRI), many existing methods remain limited by insufficient domain adaptation of pretrained models, severe class imbalance in clinical datasets, and suboptimal classification performance, which restrict their generalisability and clinical reliability. To address these challenges, this paper proposes a robust DL framework for multi-class classification of AD stages from structural MRI scans. The proposed framework is built upon a fine-tuned VGG19 model, enabling end-to-end learning and effective domain adaptation from natural images to medical imaging data, followed by a Multi-Layer Perceptron (MLP) for accurate stage-level classification. To mitigate the inherent class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is applied at the image level to balance class distributions during training. The framework is evaluated on OASIS dataset, publicly available via Kaggle, comprising 6,400 MRI images categorised into four classes. All MRI scans undergo a standardised preprocessing pipeline including image resizing, intensity normalisation, grayscale-to-RGB conversion, and training data augmentation. The dataset is split into training (80%) and testing (20%) sets using a stratified strategy. In addition, the proposed model is independently validated on ADNI dataset to assess robustness and cross-dataset generalisability. Experimental results demonstrate the effectiveness of the proposed framework, achieving an accuracy of 98.83% on OASIS and 99.13% on ADNI, along with high precision, recall, F1-score, and AUC, outperforming state-of-the-art methods.References
[1] N. Bansal and M. Parle, Dementia: an overview, Journal of Pharmaceutical Technology, Research and Management, vol. 2, no. 1, pp. 29–45, 2014.
[2] Z. Arvanitakis and D. A. Bennett, What is dementia?, JAMA, vol. 322, no. 17, 2019.
[3] C. Patterson, World Alzheimer Report 2018: The state of the art of dementia research—New frontiers, Technical Report, Alzheimer’s Disease International, Jun. 2018.
[4] S. Przedborski, M. Vila, and V. Jackson-Lewis, Series introduction: Neurodegeneration—What is it and where are we?, Journal of Clinical Investigation, vol. 111, no. 1, pp. 3–10, 2003. https://doi.org/10.1172/JCI200317522.
[5] G. M. McKhann et al., The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging–Alzheimer’s Association workgroups, Alzheimer’s & Dementia, vol. 7, no. 3, pp. 263–269, 2011.
[6] S. Sharma and K. Guleria, A deep learning model for early prediction of pneumonia using VGG19 and neural networks, in Proc. 3rd Int. Conf. Mobile Radio Communications & 5G Networks (MRCN), Springer, 2022, pp. 1–12.
[7] J. Giorgio et al., Modelling prognostic trajectories of cognitive decline due to Alzheimer’s disease, NeuroImage: Clinical, vol. 26, Art. no. 102199, 2020. https://doi.org/10.1016/j.nicl.2020.102199.
[8] C. Lian, M. Liu, J. Zhang, and D. Shen, Hierarchical fully convolutional network for joint atrophy localization and Alzheimer’s disease diagnosis using structural MRI, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 42, no. 4, pp. 880–893, 2020.
[9] E. Moser et al., Magnetic resonance imaging methodology, European Journal of Nuclear Medicine and Molecular Imaging, vol. 36, suppl. 1, pp. 30–41, 2009. https://doi.org/10.1007/s00259-008-0938-3.
[10] D. S. Marcus et al., Open Access Series of Imaging Studies (OASIS): Cross-sectional MRI data in young, middle-aged, nondemented, and demented older adults, Journal of Cognitive Neuroscience, vol. 19, no. 9, pp. 1498–1507, 2007. https://doi.org/10.1162/jocn.2007.19.9.1498.
[11] S. Kumar, K. Guleria, and S. Tiwari, An analytical study of the COVID-19 pandemic: Fatality rate and influential factors, Design Engineering, pp. 12213–12225, 2021.
[12] A. Algani et al., Machine learning in health condition check-up using Breiman’s random forest algorithm, Measurement: Sensors, vol. 23, Art. no. 100406, 2022. https://doi.org/10.1016/j.measen.2022.100406.
[13] S. Jindal et al., Artificial intelligence fuelling healthcare, in Mobile Radio Communications and 5G Networks, Springer, Singapore, pp. 501–507, 2021. https://doi.org/10.1007/978-981-15-7130-5_40.
[14] N. K. Trivedi et al., Deep learning applications on edge computing, in Machine Learning for Edge Computing, CRC Press.
[15] H. Mansourifar and W. Shi, Deep synthetic minority over-sampling technique, arXiv:2003.09788, 2020. Available: http://arxiv.org/abs/2003.09788.
[16] S. Dubey, Alzheimer’s dataset, 2019. Available: https://www.kaggle.com/tourist55/alzheimers-dataset-4-class-of-images.
[17] T. M. Ghazal and G. Issa, Alzheimer disease detection empowered with transfer learning, Computers, Materials & Continua, vol. 70, pp. 5005–5019, 2022.
[18] A. Revathi et al., Early detection of cognitive decline using machine learning and cognitive ability test, Security and Communication Networks, pp. 1–13, 2022.
[19] A. Khan and S. Zubair, Development of a three-tiered cognitive hybrid machine learning algorithm for effective diagnosis of Alzheimer’s disease, Journal of King Saud University – Computer and Information Sciences, 2022. https://doi.org/10.1016/j.jksuci.2022.07.016.
[20] D. Pirrone et al., EEG signal processing and supervised machine learning to early diagnose Alzheimer’s disease, Applied Sciences, vol. 12, no. 11, Art. no. 5413, 2022.
[21] M. Liu et al., Multi-modality cascaded convolutional neural networks for Alzheimer’s disease diagnosis, Neuroinformatics, vol. 16, nos. 3–4, pp. 295–308, 2018.
[22] D. Lu et al., Multiscale deep neural network-based analysis of FDG-PET images for early diagnosis of Alzheimer’s disease, Medical Image Analysis, vol. 46, pp. 26–34, 2018.
[23] J. Islam and Y. Zhang, Brain MRI analysis for Alzheimer’s disease diagnosis using an ensemble system of deep convolutional neural networks, Brain Informatics, vol. 5, no. 2, pp. 1–14, 2018.
[24] N. Deepa and S. P. Chokkalingam, Optimization of VGG16 using arithmetic optimization algorithm, PLoS One, vol. 15, no. 3, Art. no. e0230409, 2020.
[25] Algorithm for early detection of Alzheimer’s disease, Biomedical Signal Processing and Control, vol. 74, Art. no. 103455, 2022. https://doi.org/10.1016/j.bspc.2021.103455.
[26] S. Sharma et al., A deep learning-based CNN model with VGG16 feature extractor for Alzheimer’s disease detection using MRI scans, Measurement: Sensors, vol. 24, Art. no. 100506, 2022. https://doi.org/10.1016/j.measen.2022.100506.
[27] S. Murugan et al., DEMNET: A deep learning model for early diagnosis of Alzheimer’s disease and dementia from MR images, IEEE Access, vol. 9, pp. 90319–90329, 2021.
[28] M. Toğaçar et al., A deep feature learning model for pneumonia detection, IRBM, vol. 41, no. 4, pp. 212–222, 2020.
[29] P. Bansal et al., Improving melanoma classification accuracy using binary Harris hawks optimization, Soft Computing, vol. 26, no. 17, pp. 8163–8181, 2022.
[30] S. Gupta et al., ULBP-RF: A hybrid approach for malware image classification, in Proc. 5th Int. Conf. Parallel, Distributed and Grid Computing (PDGC), 2018.
[31] K. Simonyan and A. Zisserman, Very deep convolutional networks for large-scale image recognition, arXiv:1409.1556, 2014.
[32] J. Pardede et al., Transfer learning using VGG16 for fruit ripeness detection, International Journal of Intelligent Systems and Applications, vol. 13, no. 2, pp. 52–61, 2021.
[33] P. Xu et al., Identification of intrinsically disordered protein regions using deep neural networks, Algorithms, vol. 14, no. 4, Art. no. 107, 2021.
[34] A. De and A. S. Chowdhury, DTI-based Alzheimer’s disease classification with rank-modulated fusion of CNNs and random forest, Expert Systems with Applications, vol. 169, Art. no. 114338, 2021.
[35] E. K. Fanar et al., Diagnosis of Alzheimer’s disease using 2D MRI slices by CNN, Applied Bionics and Biomechanics, Art. no. 6690539, 2021.
[36] T. Pan et al., Learning imbalanced datasets based on SMOTE and Gaussian distribution, Information Sciences, vol. 512, pp. 1214–1233, 2020.
[37] S. Ahmed et al., Ensembles of patch-based classifiers for Alzheimer’s disease diagnosis, IEEE Access, vol. 7, pp. 73373–73383, 2019.
[38] D. Stamate et al., Applying deep learning to predicting dementia and mild cognitive impairment, in Artificial Intelligence Applications and Innovations, IFIP AICT, vol. 584, Springer, 2020, pp. 308–319.
[39] R. C. Petersen et al., Alzheimer’s Disease Neuroimaging Initiative (ADNI): Clinical characterization, Neurology, vol. 74, no. 3, pp. 201–209, 2010.
DOI:
https://doi.org/10.31449/inf.v50i9.8187Keywords:
alzheimer’s disease, image analysis, deep learning, fine-tuning, MLP, VGG19, MRIDownloads
Published
How to Cite
Issue
Section
License
Authors retain copyright in their work. By submitting to and publishing with Informatica, authors grant the publisher (Slovene Society Informatika) the non-exclusive right to publish, reproduce, and distribute the article and to identify itself as the original publisher.
All articles are published under the Creative Commons Attribution license CC BY 3.0. Under this license, others may share and adapt the work for any purpose, provided appropriate credit is given and changes (if any) are indicated.
Authors may deposit and share the submitted version, accepted manuscript, and published version, provided the original publication in Informatica is properly cited.
