Utilizing the Random Forest Algorithm to Enhance Alzheimer’s disease Diagnosis

Abstract

Machine learning is widely used in many aspects of healthcare. The development of medical technology has made it possible to gather better data for early disease symptom diagnosis. This study makes an effort to categorize Alzheimer’s disorder. Alzheimer’s disease is a fatal disorder that may result in memory loss and mental impairment. To prepare for medical attention, this needs early disease diagnosis. Magnetic resonance imaging (MRI) can be used to accurately and non-invasively diagnose Alzheimer’s disease. Effective feature extraction and segmentation techniques are necessary for the accurate diagnosis of MRI images. Utilizing MRI data of the brain’s white matter, grey matter, and cerebrospinal fluid, feature selection is carried out. Random forest trees are used in standard machine learning methods like regression and classification. The results of the utilized method were next contrasted with those of other machine learning techniques. As a result, RF model-based interpolation analysis surpasses the RF non-imputation method with greater accuracy, specificity, sensitivity, f-measure, and ROC.