Disease Prediction using Machine Learning Classification Algorithms

Abstract

Alzheimer's disease is the main cause of dementia in elderly people. The current generation, a great deal of interest in applying machine learning to explore metabolic diseases that affect large numbers of people worldwide, such as Alzheimer's disease and diabetes. Its incidence is increasing at an alarming rate each year. Neurodegenerative changes affect the brain in Alzheimer's disease. As the population ages, more and more people, their families, and health care workers will suffer from diseases that affect memory and function. The impact is severe socially, financially and economically. Alzheimer's disease is difficult to predict in its early stages. Treatment in the early stages of Alzheimer's disease is more effective and causes less damage than treatment in the later stages. Several techniques such as decision trees, random forests, support vector machines, XGBoost, extra-tree classifiers, gradient boosting, AdaBoost, and voting classifiers were used to identify the optimal parameters for predicting Alzheimer's disease. After this, the dataset was trained with Neural Networks to depict that neural nets are not recommended with small datasets. These predictions of the disease are based on Open Access Series of Imaging Studies (OASIS) data, and performance is measured using parameters such as ML model accuracy, recall, accuracy, and F1 score.