Performance Evaluation of Deep Learning Models for Alzheimer’s Disease Detection

Abstract

Alzheimer's disease (AD) presents as a neurodegenerative condition characterized by dementia, impacting cognitive function, memory, and behavior. Swift identification is crucial for effective intervention and better patient outcomes. This research aims to tackle this urgent need by harnessing deep learning techniques to advance early AD detection. By merging magnetic resonance imaging (MRI) with advanced deep learning algorithms, the objective is to create a precise, non-invasive approach for early AD diagnosis. The primary goal is to uncover new features and significantly boost the accuracy of AD detection. The method involves examining convolutional neural networks alongside MRI scans to reveal patterns indicative of AD progression. Notably, the proposed model achieves impressive evaluation metrics: 99.19% accuracy, 0.023% loss, 99.08% f1-score, and 99.11% precision. This endeavor seeks to demonstrate substantial advancements in AD detection accuracy compared to existing methods. Through this interdisciplinary approach, the aim is to drive progress in early AD diagnosis, ultimately leading to more effective interventions and enhanced quality of life for those impacted by this challenging condition accuracy compared to existing methods.