Classification and Investigation of Alzheimer Disease Using Machine Learning Algorithms

Dementia is a globally identified problem. The occurrence of dementia increases abruptly with growing age.It is an irreversible brain disease which causes degeneration in the cognitive ability of a person affecting his thinking, memory and judgment. Throughout the world around 50 million people have dementia and around 10 million new cases are diagnosed every year. Hence addressing this issue has become need of the hour and early diagnosis of dementia is essential for the progress of more prevailing treatments. Early diagnosis of this disease is done using cognitive tests to determine the mental ability of a person. Some of the cognitive tests include CDR, MMSE, and Adden Brooke's cognitive examination. In present research work using machine learning techniques we have tried to detect the dementia in early stage.The data composed for investigation consists of the gender, age education,MMSE,CDR,ASF,Handedness,number of visits of the patient to the hospital who are clustered as demented or non-demented. We have used different machine learning algorithms like Random forest classifier, (SVM), Decision Tree Classifier, Extra Tree Classifier, Neighbors Classifier and Logistic Regression to analyze the data.The comparison study of each algorithm is done. The algorithm with highest accuracy will be used to further data analysis. In our proposed work we have used extra tree classifier is used for more examination of the facts.