Alzheimer's Disease Detection Using Ensemble Learning and Artificial Neural Networks

This paper presents an ensemble method using machine learning classification algorithms and an artificial neural network-based scheme using the popular and widely used open access series of imaging studies (OASIS) dataset for Alzheimer's disease (AD) detection. The proposed work performs an in-depth feature examination and a training-test split in a 70 : 30 ratio on the dataset and applies 8 different ML algorithms. The AD detection outcome is obtained using two procedures, first by an ensemble approach applied to different machine learning algorithms, and secondly by using an artificial neural network (ANN). The use of ANN achieves an overall test accuracy of 0.9196 whereas two ensemble techniques, namely gradient boosting and voting classifier achieve an overall test accuracy of 0.857 and 0.8304. The precision and sensitivity scores demonstrate the superior detection performance of the ANN over the ensemble method on ML algorithms.