eng Oriental Scientific Publishing Company Biomedical and Pharmacology Journal 0974-6242 2025-11-04 18 October Spl Edition 68424 article Upendra Singh 1 Vidit Kumar 1 Bhaskar Pant 1 Department of Computer Science and Engineering, Graphic Era (Deemed to be University), Dehradun, India, Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder associated with irreversible structural and functional brain deterioration. Early and accurate diagnosis is essential to slow disease progression and improve patient quality of life. Existing methods face several challenges, including limited ability to capture multi-scale volumetric features, reliance on handcrafted or shallow feature representations, and insufficient focus on disease-relevant brain regions, which often reduce diagnostic reliability. To address these limitations, this work aims to develop an effective computer-aided diagnostic framework based on 3D CNN and attention using magnetic resonance imaging (MRI) to distinguish between AD, mild cognitive impairment (MCI), and cognitively normal (CN) subjects. We proposed Alzh-3DCNN-Attention, a modified C3D model that integrates a Convolutional Block Attention Module (CBAM) with sequentially connected 3D CNN layers to emphasize discriminative brain regions while preserving volumetric context. The novelty of our approach lies in its ability to jointly capture multi-scale features and focus on disease-relevant regions without the need for handcrafted feature engineering. Experiments were conducted on the ADNI dataset with rigorous pre-processing to enhance data quality. The proposed model achieved 91.54% accuracy, 89.75% sensitivity, 92.35% specificity, and 89.70% AUC for multi-class classification, outperforming conventional 3D CNN baselines. These results demonstrate that attention-enhanced 3D CNNs improve feature representation and yield robust diagnostic performance. The findings suggest that the proposed approach provides a reliable and efficient tool for automated AD classification, offering valuable support for clinicians in early disease detection. https://biomedpharmajournal.org/vol18octoberspledition/alzheimers-disease-classification-usingattention-based-3d-convolutional-neural-network-and-3-d-mri-images/ Convolutional-3D Cognitively normal 3-D MRI Deep learning Image-classification Mild cognitive impairment