Developing Hybrid Machine Learning Models for Early Dementia Detection by Utilising Genetic and Neuroimaging Data

Dementia, especially Alzheimer’s disease (AD), is a neurological condition that causes behavioural abnormalities, memory loss, and cognitive decline. Early diagnosis is essential, and new developments in artificial intelligence (AI) and magnetic resonance imaging (MRI) have improved diagnostic techniques. Using genetic and neuroimaging data, this work examines hybrid models of machine learning— Support Vector Machines (SVM) and Random Forest (RF) in particular—for dementia stage classification. To improve the accuracy and resilience of the model, methods such as feature engineering and data augmentation are applied. Problems including algorithm optimisation, data asymmetry, and clinical adaptation still exist despite encouraging outcomes. This study underlines the need for more advancements to increase clinical applicability while highlighting the potential of hybrid AI models to revolutionise dementia diagnosis.