Preliminary Detection of Dementia using Machine Learning Techniques

Dementia can lead to a patient's memory loss and language impairment. Dementia is rising day by day and it is one of the most common reasons for deaths in recent years. It is a continuing challenge of the 21st century and an unnoticed issue that will become more concerning if negligence continues in the future, and it can also affect a large number of the population if keeps increasing at the present pace. Detecting dementia is a major and time-consuming task. To overcome this issue of dementia detection we have tried to figure out a preliminary and feasible approach to detect dementia, which takes basic attributes that can be simply tracked by using a smartwatch and general health equipment. We have taken the data from Kaggle which has 8 attributes and 3000 rows. We have analyzed the data using Machine Learning Methods and Achieved a 79% accuracy level. The study done in this paper indicates that deep learning and machine learning can be utilised to identify dementia at early stages at an affordable price range with less time consumption and can aligned with daily activities to get more accurate and precise results in the future.