Feature Selection as a Tool to Support the Diagnosis of Cognitive Impairments Through Handwriting Analysis

Abstract

Cognitive Impairments are cognitive deficits that are greater than expected for a person of a given age and level of education, but which do not significantly interfere with the daily life of the people affected. They range from mild to severe and are seen as a risk factor for Alzheimer’s disease, currently the most common neurodegenerative brain disorder worldwide. In a previous study, we presented an experimental protocol comprising different handwriting tasks to be carried out by patients and a healthy control group: the aim was to investigate whether the analysis of the handwriting could be used as a tool to support the diagnosis of this kind of impairment. In the study presented here, we used a well-known and widely-used feature selection approach to determine the most effective features for predicting the symptoms related to cognitive impairments via handwriting analysis. Our intention is to deepen the knowledge about the different cognitive functions affected by the onset of these diseases, as well as to improve the performance of the tools developed to support their diagnosis. The results showed that different sets of highly discriminant features, closely related to the cognitive skills impaired, were selected for the handwriting tasks making up the protocol, thus supporting our hypothesis that their use can be very helpful to support the diagnosis of cognitive impairment.

Keywords

• Medical expert systems
• cognitive impairments
• feature selection