A Hybrid Approach for Recognising Alzheimer Disease Using GA Enabled GAN

Alzheimer's disease (AD) is a chronic neurodegenerative disease, where early stages of detection play a crucial role in improving patient treatment. However, patients with Fronto Temporal Dementia (FTD) are often misdiagnosed with Alzheimer's disease due to limitations in diagnosis algorithms. While various machine learning and deep learning techniques have been applied to classify the stages of Alzheimer's disease, accurate early-stage diagnosis remains challenging which would give a better explanation of its variations with FTD. Though many theories have been developed to diagnose it, Magnetic Resonance Imaging (MRI) has proven to be a powerful tool in observing detailed patient images. Several machine learning models have been exploited by researchers to diagnose Alzheimer's disease. Though many techniques have been applied to the diagnosis and classification of Alzheimer's disease, there is a need for more accuracy in early diagnosis solutions. In this study, the Open Access Series of Imaging Studies (OASIS) dataset is used to predict the disease diagnosis using the Genetic Algorithm (GA). The results of GAGAN (Genetic Algorithm based Generative Adversarial Network) model is compared to traditional GAN (Generative Adversarial Network) to evaluate performance, with a specific focus on improved classification accuracy and sensitivity. Results demonstrate that GA optimization enhances GAN's diagnostic accuracy, offering a more robust framework for real-world application in early AD detection, which can contribute to better patient care and quality of life.