To search, Click below search items.

 

All Published Papers Search Service

Title

Predicting the Relation between Education Level, Age and Alzheimer¡¯s Disease by Simple Linear Regression

Author

Samah Abuayeid, Hosam Alhakami, Abdullah Baz, Tahani Alsubait

Citation

Vol. 20  No. 4  pp. 191-200

Abstract

Alzheimer¡¯s disease (AD) is a progressive perturbation of brain cells that could cause decay in human social behavior. Moreover, AD is considered the most common reason for dementia that affects many people in old age. There is a demand for understanding AD risk factors to decrease the possibility of AD. These factors are classified into two categories: modified factors (lifestyle and education) and unmodified factors (age, gender, and genetic). Data analytics and machine learning techniques have been introduced in bioinformatics research to diagnose, predict, and prevent AD. The primary aim of this research is to study the relationship between education level and AD, where education level is considered as a modified risk factor. Additionally, to improve our results, we also study the relationship between AD and age as an unmodified risk factor, and then we compare the two sets of results. We have used a publicly available dataset from the Kaggle community. We have built two linear regression models where the first studies the relation between AD and education level and the second studies the relationship between patients¡¯ age and AD. We have observed that education level affects AD patients negatively if only multiple risk factors are available in the patient¡¯s environment, and that will increase the possibility of AD conversion to dementia. In general, the results demonstrate the ability of linear regression in predicting that the combination of AD risk factors could affect AD people negatively. These results encourage taking measures to enhance the AD patient¡¯s environment and reduce the number of risk factors.

Keywords

Alzheimer¡¯s disease Dementia MCI False memory Machine learning data analysis Linear regression.

URL

http://paper.ijcsns.org/07_book/202004/20200424.pdf