Research Article
Semi-Supervised GAN Driven Feature Augmentation for Alzheimer Disease Detection
@INPROCEEDINGS{10.4108/eai.28-4-2025.2358109, author={V. Asritha and Siddhhika Subramanian and K. Vasavi and Azhagiri Mahendiran}, title={Semi-Supervised GAN Driven Feature Augmentation for Alzheimer Disease Detection}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part II}, publisher={EAI}, proceedings_a={ICITSM PART II}, year={2025}, month={10}, keywords={alzheimer’s disease detection semi-supervised learning generative adversarial networks (gan) feature augmentation mri classification}, doi={10.4108/eai.28-4-2025.2358109} }- V. Asritha
Siddhhika Subramanian
K. Vasavi
Azhagiri Mahendiran
Year: 2025
Semi-Supervised GAN Driven Feature Augmentation for Alzheimer Disease Detection
ICITSM PART II
EAI
DOI: 10.4108/eai.28-4-2025.2358109
Abstract
Alzheimer’s Disease (AD) is a progressive neuro- degenerative disorder that early and accurate diagnosis is an important task to ensure the efficiency of the treatment. But the absence of annotated medical imaging data is a bottleneck for the development of accurate AI models for Alzheimer's detection. This adversarial GAN-based feature augmentation method has the potential to improve the classification performance of Alzheimer's disease. The fake and realistic brain imaging features generated by the GAN model enhance the small labelled dataset. A semi-supervised learning is, on the other hand, involving labelled and unlabelled data to enhance the feature representations. Our model achieved the following on the benchmark AD datasets compared to traditional supervised learning methods (Table 6): AUC-ROC of 95.2%, accuracy of 96.5%, precision of 95.8%, recall of 96.2% and F1-score of 96.0%. These findings indicate promise toward GAN-mediated feature expansion into radiology imaging, for early accurate diagnosis of Alzheimer’s disease.
