• Tejaswini
• Nikhil Murade
• Dr. M.A. Wakchaure

Alzheimer's disease (AD) is a progressive neurodegenerative disorder that severely impacts cognitive function and quality of life. Early and accurate diagnosis is crucial for effective management and treatment. This research presents a deep learning-based approach for the early detection of Alzheimer's disease using brain imaging data, focusing on creating lightweight models optimized for deployment on resource-constrained devices. Leveraging transfer learning with pre-trained models (MobileNetV2, DenseNet121, and ResNet50) and applying post-training quantization techniques, we significantly reduce model size and computational requirements without sacrificing accuracy. Rigorous experimentation on an augmented Alzheimer's MRI dataset demonstrates that quantized models achieve high classification performance, maintaining over 85% accuracy while reducing model size by up to 77.5%. Furthermore, an ensemble learning strategy combining the three models yields a notable improvement, achieving 91.56% accuracy even in its quantized form. These findings highlight the potential of lightweight deep learning solutions for accessible, real-time Alzheimer's disease detection, particularly in remote and resource-limited healthcare settings.

Deep Learning, Alzheimer's Disease Detection, Transfer Learning, Quantization, MobileNetV2, DenseNet121, ResNet50, Ensemble Learning, Edge Computing in Healthcare

"A DEEP LEARNING APPROACH FOR EARLY DIAGNOSIS OF ALZHEIMER'S DISEASE USING BRAIN IMAGING DATA", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.n242-n250, April-2025, Available :http://www.jetir.org/papers/JETIR2504D32.pdf

"A DEEP LEARNING APPROACH FOR EARLY DIAGNOSIS OF ALZHEIMER'S DISEASE USING BRAIN IMAGING DATA", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppn242-n250, April-2025, Available at : http://www.jetir.org/papers/JETIR2504D32.pdf