A literature review of online handwriting analysis to detect Parkinson’s disease at an early stage

Parkinson’s disease (PD) affects millions of people worldwide, it dramatically affects the brain areas’ structure and functions. Therefore, it causes a progressive decline of cognitive, functional and behavioral abilities. These changes in the brain result in the degradation of motor skills’ performances. Handwriting is a daily task combining cognitive, kinesthetic and perceptual-motor abilities. Thus, any change in the brain areas affects directly on the aspects of handwriting. For this purpose, many researchers have studied the possibility of using the handwriting alterations caused by PD as diagnostic signs, in order to develop an autonomic and reliable Diagnosis Aid System which could strongly detect this pathology at an early stage. This intelligent system could help in assessing and controlling the evolution of PD, and consequently, in the improvement of the patients’ quality of life. This paper aims at presenting a literature review of the most relevant studies conducted in the area of the on line handwriting analysis, in order to support PD. Starting by the typical followed procedure which consists of handwriting data acquisition, used materiel, proposed tasks, feature extraction, and finally data analysis. Indeed, according to all the investigated studies, dynamic handwriting analysis is a powerful, noninvasive, and low-cost tool to effectively diagnosis PD. In conclusion of the paper, future directions and open issues are highlighted.