Streamlining Deep Learning Based Alzheimers Disease Classification: A Simplified Approach with ADASYN

Consuming and complex, while existing deep learning (DL) methods may suffer from complexity and overfitting issues. This study proposes an integrated approach that combines Convolutional Neural Networks (CNNs) with the Adaptive Synthetic Sampling (ADASYN) algorithm to address class imbalance and enhance AD classification accuracy. This model architecture prioritizes simplicity and efficiency, utilizing an optimal number of layers to achieve effectiveness. Experimental results on a dataset of brain MRI images demonstrate the efficacy of this approach, with a classification accuracy of 98%. By integrating machine learning techniques with advanced sampling strategies, this research contributes to advancing early detection and diagnosis of Alzheimer's Disease, fostering strides towards improved patient care and management.