Motor patterns recognition in parkinson's disease

Abstract

Parkinson's disease (PD) is characterized clinically by main motor symptoms such as tremor at rest, rigidity, and bradykinesia that affect movements, including gait and postural adjustments. The diagnosis is based on the clinical recognition of these symptoms with the consequent high interrater variability. In order to perform an objective and early diagnosis, approaches that overcome the limitations inherent to clinical examination are needed. In the present work, we will describe several classical technological approaches, such as 3D motion analysis, to achieve an objective evaluation of the cardinal motor symptoms in PD. Furthermore, we will take into account the attempts to identify pathological patterns of integrated, more complex functions such as gait and posture. Finally, as future directions, we will discuss the machine learning approaches in the individuation of specific gait patterns in PD.