21 Conditional entropy-based evaluation of the postural control complexity in patients with Parkinson’s disease

Abstract

The Parkinson’s disease (PD) patient exhibits postural instability and difficulty in generating corrective action. The aim of this study was to characterize postural control of PD patients and to describe the effects of electrical stimulation of sub-cortical areas. The posture of 12 patients with PD undergoing deep brain stimulation (DBS) was studied during eyes open (EO) and eyes closed (EC) before surgery, after electrodes implantation and after device activation. This study proposes a new method to quantify the postural control based on the complexity analysis of the signal representing the sway of the center of pressure (CoP) during a 30 s orthostatic standing in sagittal and in frontal plane. The signal complexity was quantified by the conditional entropy: an higher conditional entropy means a lower predictability and a greater complexity of the postural control. A temporal and spectral analysis of the CoP series were also performed. Statistical analysis showed a significant influence of the direction of sway and of the visual control on the complexity of postural control and on both temporal and spectral characteristics of the CoP series. A significant influence of the electrical stimulation on both temporal and spectral analysis but not on the complexity of postural control was also found. This finding could be justified by a reduction of muscle stiffness after surgical intervention. The lack of visual control may simplify postural control connectivity due to an increased required attention level. A short term sub-cortical stimulation has no effects on PD postural strategy.