A Logistic Regression and Decision Tree Based Hybrid Approach to Predict Alzheimer's Disease

Alzheimer's disease is a degenerative neurological disorder that typically impacts individuals over the age of 65, causing damage to the brain and resulting in challenges with memory, cognition, and behavior. Although there is currently no cure for this condition, various therapies and medications exist to manage symptoms and slow the progression of the disease. Individuals with Alzheimer's disease experience cognitive impairment, changes in behavior, and communication difficulties, often leading to psychological and behavioral issues like anxiety, aggression, and depression. In this research paper, a novel approach to predicting outcomes using MRI data sets from the Open Access Series of Imaging Studies (OASIS) is proposed. The method involves data exploration, data preprocessing, and the development of a hybrid model that utilizes both Logistic Regression and Decision Tree algorithms. Through rigorous testing, the proposed hybrid model demonstrated superior accuracy compared to existing models, achieving an overall accuracy rate of 96%.