Machine Learning Approach for Early Detection of Alzheimer’s Disease

Abstract

Alzheimer’s Disease (AD) is thought to be the most generally known cause of dementia, and it is estimated that only one in four people with Alzheimer’s are properly identified. While there is no definitive cure, the negative effects may be ignored while the weakening is mild. Treatment is most effective when it begins before major downstream injury occurs, i.e., at the stage of moderate cognitive impairment (MCI) or much earlier. Physical and neurological examinations and neuropsychological and cognitive testing are used to diagnose Alzheimer’s disease. There is a need for improved diagnostic tools, which is what this postulation addresses. Kaggle is an online open-access dataset for improving the Alzheimer’s disease diagnosis technique. The information acquired during the conference is documented. One goal of this theory’s research is to look at machine learning methodologies to create a classifier that can help screen new persons for different stages of Alzheimer’s disease. In comparison to previous work methods, our methodology is suitable for breaking down diverse classes in a single setting and needs less distinct training samples and inconsequential prior knowledge. In our tests, we saw a significant improvement in categorizing all diagnostic categories. Initially, the model is trained on 64 evaluated examples from the Kaggle database. We next test our created model on the whole set of entries supplied by the Kaggle dataset to confirm our framework’s finding. Our results show 96.37% accuracy in AD detection and categorization.