Machine Learning Based Alzheimer’s Disease Detection: A Comprehensive Approach

Alzheimer’s disease typically leads in the death of neurons and the connections between them in memory-related parts of the brain, such as the entorhinal cortex and hippocampus. Following that, it affects the areas of the cerebral cortex responsible for language, cognition, and social interaction. Several other regions of the brain get harmed at some time. Patients with Alzheimer’s disease gradually lose their capacity to function normally and go about their daily lives. Currently, age is regarded as one of the most important risk factors; the majority of patients with dementia are 65 or older. Symptoms of the condition include increased disorientation, memory loss, including the inability to recall recent events or personal history, and learning difficulties. As of yet, there is no treatment for Alzheimer’s. A vast quantity of data will be acquired from the previously affected person, and the user input data will be compared to past data using machine learning techniques to determine whether they are still impacted. Finding the best categorization approach is difficult because it necessitates adapting to regional datasets. The methods LightGBM, AdaBoost, Random Forest Classifier, Support-Vector-Machine, Decision-Tree, and XGBoost provided accuracy rates of 78%, 86%, 81%, 73%, and 84%, respectively. In this work, the Voting Classifier algorithm, which may be more accurate than other methods, will be employed.