Alzheimer’s Disease Detection Based on Various Subjects Provided Through Comprehensive Diagnosis and MRI Imaging Using Machine Learning Techniques

With the rapid growth of technology and innovations, many have shown considerable progress in assisting in the health and medical field. Machine learning techniques are significantly used in detecting various predictable diseases. In this proposed study of Alzheimer’s disease (AD), it is detected using various machine learning techniques. The objective of this paper is to categorize the disease into two one is demented and another is non-demented based on the various parameters. This classification can be achieved using analysis of significant details of the brain that are provided through various brain examinations including Magnetic Resonance Imaging (MRI) scans. Machine learning algorithms used in this research to predict dementia are Support vector machine (SVM), Decision tree, Logistic regression, and Random forest classifier. Also, we have used ensemble boosting techniques such as AdaBoost and XGBoost. Among all the techniques used, the evaluating performance of the XGBoost boosting classifier has delivered the most promising results in providing the most accurate predictions of the test data. The percentage of accuracy we got for this model is 95%. Values of other evaluating parameters are recorded as follows: for precision we got 95%, for recall we got 98%, and the F1-score was 96%. In the future deep learning techniques can be employed on the dataset containing various Alzheimer’s features and can provide more improved results in Alzheimer’s prediction.