Early Onset Detection of Alzheimer's Disease Based on Intelligent Techniques

Abstract

Alzheimer's disease (AD) is a life-threatening disease in senior citizens. Alzheimer's disease affects reasoning and recollection while also causing the overall size of the brain to diminish, ultimately leading to death. The development of more effective therapies for AD depends on an early identification of the condition. In this chapter, authors propose to use machine learning techniques for early onset detection of AD. Authors have generated a dataset based on features which represent the early symptoms of AD. Experimental results have been obtained using Random Forest, SVM, XGBoost, and Naive Bayes classifiers. The experimental results have been evaluated using metrics such as the confusion matrix, accuracy, and sensitivity. The XGBoost model provides an average validation accuracy of 86% on AD test data which is comparable to the well-established techniques in the literature.