Classification of Patients with the Development of Alzheimer's Disease using an Ensemble of Machine Learning Models

Every year, the number of diagnosed cases of Alzheimer's disease (AD) continues to grow. Dementia affects memory, orientation, language, learning ability, and the ability to perform daily activities. It is very important to correctly diagnose the stage of Alzheimer's disease, as each stage requires different treatment and support strategies for the person and their caregivers. Machine learning (ML) methods have been shown to be effective in the classification of AD patients based on medical images, such as magnetic resonance imaging (MRI). However, individual ML models often have limited performance due to overfitting or the inability to capture all of the complex patterns in the data. In this study, an ensemble of ML models is proposed to improve the classification of patients with the development of AD. The ensemble model combines the predictions of multiple individual ML models, such as Random Forest, Multi-Layer Perceptron and SVM, to produce a more accurate and robust prediction. The ensemble model achieved an accuracy of 96% in classifying patients into five stages of AD: cognitively normal, early mild cognitive impairment, late mild cognitive impairment, mild cognitive impairment, and Alzheimer's dementia.