Visual Perturbation Suggests Increased Effort to Maintain Balance in Early Stages of Parkinson’s to be an Effect of Age Rather Than Disease

Abstract

Postural instability marks a prevalent symptom of Parkinson’s disease (PD). It often manifests in increased body sway, which is commonly assessed by tracking the Center of Pressure (CoP). Yet, in terms of postural control, the body’s Center of Mass (CoM), and not CoP is what is regulated in a gravitational field. The aim of this study was to explore the effect of early- to mid-stage PD on these measures of postural control in response to unpredictable visual perturbations. We investigated three cohorts: (i) 18 patients with early to mid-stage PD [Hoehn & Yahr stage (1–3); 1.94 ± 0.70]; (ii) a group of 15 age-matched controls (ECT); and (iii) a group of 12 young healthy adults (YCT). Participants stood on a force plate to track their CoP, while the movement of their entire body was recorded with a video-based motion tracking system to monitor their CoM. A moving room paradigm was applied through a head-mounted virtual reality headset. The stimulus consisted of a virtual tunnel that stretched in the anterior-posterior direction which either remained static or moved back and forth in an unpredictable fashion.We found differences in mean sway amplitude (MSA) and mean velocities of CoP and CoM between the groups under both conditions, with higher MSA of CoP and CoM for PD and higher mean velocities of both variables for PD and ECT when compared with YCT. Visual perturbation increased mean CoP velocity in all groups but did not have effects on mean CoM velocity or MSA. While being significantly lower for the young adults, the net effect of visual perturbation on mean CoP velocity was similar between patients with PD and age-matched controls. There was no effect of the visual perturbation on mean CoM velocity for any of the groups.Our simultaneous assessment of CoP and CoM revealed that postural control is reflected differently in CoM and CoP. As the motion of CoM remained mostly unaffected, all groups successfully counteracted the perturbation and maintained their balance. Higher CoP velocity for PD and ECT revealed increased corrective motion needed to achieve this, which however was similar in both groups. Thus, our results suggest increased effort, expressed in CoP velocity, to be an effect of age rather than disease in earlier stages of PD.

Convolutional neural network reveals frequency content of medio-lateral COM body sway to be highly predictive of Parkinson’s disease

“…However, investigating dynamic balance scenarios usually requires much higher technical and experimental effort. Our models were not better at distinguishing between the groups based on data taken from the dynamic conditions of our previous studies (Engel, Student, et al, 2021;Student et al, 2022). This is remarkable, since it renders the bulk of the setup obsolete.…”

“…All individuals with PD were measured while being "on" their regular dose of dopaminergic medication (Levodopa Equivalent Daily Dose (LEDD): ; 651.63 ± 529.97). Detailed information on the individuals recruited for our PD group can be found in our previous work (Student et al, 2022). For the control group (healthy controls, HC), we recruited fifteen age-matched healthy adults (age: [49-70]; 59.80 ± 6.45).…”

“…Given the persisting lack of suitable objective measures to assist clinical balance assessment in PD and the promising application of machine learning methods to various aspects of the disease, the aim of our study was to develop a CNN classifier trained on short excerpts of static posturographic data that will be able to generalize for new subjects based on a quick and easy assessment. For this purpose, we recorded the COP and COM trajectories of individuals with PD and age-matched HC during quiet standing using mobile and cost-effective devices Engel, Student, et al, 2021;Student et al, 2022) and trained a custom-built CNN to distinguish both groups based on the time-frequency content represented as spectrogram images. Moreover, along the urgent need of explainable AI, we used a GradCAM approach (Selvaraju et al, 2020) to understand which frequencies in the postural sway data were vital to the classifier's decision.…”

Balance Assessment in Selected Stages of Parkinson’s Disease Using Stock Exchange Indices

“…The comparison of the CGy group with the CG group clearly demonstrates that in the CG group, there was a statistically signi cant prolongation of time (TCI_dT) in the EO and EC tests as well as the distance (TCI_dS) in the EO tests. This prolongation of time, i.e., the decrease in subsequent postural corrections, re ects the changing balance-keeping strategy and may indicate the declining role of the proprioceptive system in maintaining balance in favor of the organ of sight [42,43].…”

“…Such an extension of the TCI_dS distance entailing an increase in the velocity of leaps of the COP to subsequent positions (TCI_dV) may also be an indication of a greater probability of the fall occurrence. Moreover, an unpredictable destabilizing factor combined with an increased value of the COP velocity may lead to such an increase in the velocity vector value that the deceleration of the body movement becomes impossible and results in the crossing of the stability boundary [41,42].…”