Comparative Study of Dimensionality Reduction Techniques and Machine Learning Algorithms for Alzheimer's Disease Classification and Prediction

Dementia is one of the most serious diseases affecting the elderly population. It is distinguished by several symptoms that affect the memory and the normal life of the patients. According to the most recent statistics, almost 60% to 80% of people diagnosed as suffering from dementia are due to Alzheimer's disease. The number of diagnosed patients is increasing year after year, which is why it has become a significant world health issue, deserving more and more focus and attention.This paper presents a comparative overview of dimensionality reduction techniques and machine learning algorithms. The used dataset is a cross-sectional of 416 cases from a collection of 434 MRI sessions.In this study, we will use different dimensionality reduction techniques, machine learning algorithms, data preprocessing transformations, and feature selection methods to classify the input data for early prediction of dementia in suspected subjects. The best predictions are around 86% for different algorithms, this will allow improving the classification of Alzheimer's disease based on the finest machine learning algorithms, which in its turn would allow researchers to find better ways to slow or even stop the disease process in its first stages.