1066 Cortico-Pallidal Recordings During Walking Identify Neural Features Predictive of Impaired Turning in Patients With Parkinson’s Disease
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
INTRODUCTION: Turning is a core aspect of everyday walking which is often impaired in Parkinson’s Disease (PD), contributing to falls and significant morbidity. The neural control of turning and its dysfunction in PD are poorly understood. METHODS: Field potentials were streamed form the globus pallidus internus (GPi) and externus (GPe), primary motor cortex (M1), and premotor cortex (PMC) from five PD patients with gait impairments using a bidirectional neural interface during an overground walking task. Kinematic measurements were obtained from body-worn IMUs. Tasks were performed during on and low medication states, and with DBS turned on and off. RESULTS: Cortical beta power decreased significantly during turning compared to walking (p < 0.01). Using a novel turning performance index (TPI) validated based on clinician ratings, we identified patients whose turning improved with medication state as well as stimulation state. Pallidal beta power was significantly reduced with medication among improvers, as was GPe-M1 beta coherence (p < 0.001), but no difference was seen for medication non-responders. Stimulation-associated improvements in turning is marked by decreased M1 beta power while non-improver showed higher M1 beta power (p < 0.001). Beta bursting analysis revealed higher burst count in all brain regions during poor turns, and burst amplitude inversely correlated with TPI. Linear discriminant analysis classifiers were trained on neural data and resulted in AUC values of 0.57-0.81 in discriminating turns from straight-line walking. R2 values for predicting turning performance using various machine learning models ranged from 0.32-0.71. CONCLUSIONS: Impaired turning is associated with increased beta power both in the pallidum and cortex, with improvements seen when these oscillations are reduced with medications or stimulation. Robust patient-specific classifiers to detect turns and regression models to estimate turning performance can be trained from field potential recordings.
Fields of Study & Subjects
Deep Brain Stimulation and Movement Disorders
Balance
Gait
and Falls Prevention
Motor Control and Adaptation
Publication details
- Journal / source
- Neurosurgery
- Volume
- 72
- Issue
- Supplement_1
- Pages
- 194–195
- Publication date
- 26 March 2026
- Language
- EN