Characterizing cognitive impairment through drawing features extracted from the Tree Drawing Test

Abstract

Dementia affects millions of people around the world and is still growing rapidly. Mild cognitive impairment(MCI) and Alzheimer’s disease(AD) are prodromal symptoms leading to dementia. Therefore, it is necessary to distinguish MCI and AD from healthy middle-aged and elderly people. Many studies have shown that people with dementia have difficulty drawing. Complicated drawing tasks (such as the Tree Drawing Test) can better observe drawing obstacles than copying tasks. Previous studies have used the Tree Drawing Test(TDT) to screen for cognitive impairment, but few of them have focused and comprehensively paid attention to the details of tree drawing. In this study, we recruited 81 subjects and collected their tree drawing data via a digital tablet. Starting from two feature dimensions, quantitative and qualitative, we extracted 51 features from tree drawing details and globals. We conducted 5 sets of binary classification experiments using widely used machine learning models and additional experiments to eliminate the interference of gender, education, and age. We then ranked the importance of the features selected by the model. It can be seen from the machine learning classification results that the binary classification of different cognitive situations has good performance, and the features we extracted play a great role. This preliminary study suggests that the TDT has the potential to be used to develop machine learning tools to support the diagnosis of cognitive impairment disorders.