🧠 neuroscience

Reduced entropy of subthalamic beta bursts predicts freezing of gait in Parkinsons disease

This study demonstrates that reduced entropy in the timing of subthalamic nucleus beta bursts precedes and predicts the onset of freezing of gait in Parkinson's disease, indicating a transition to constrained neural dynamics.

Original authors: Beaudoin, C. A., OKeeffe, A. B., Abdi-Sargezeh, B., Gillies, M. J., Oswal, A., Green, A. L.

Published 2026-08-21

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Beaudoin, C. A., OKeeffe, A. B., Abdi-Sargezeh, B., Gillies, M. J., Oswal, A., Green, A. L.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

In the human brain, deep within a structure called the subthalamic nucleus, tiny electrical signals fire in rhythmic patterns that help coordinate movement. In a condition known as Parkinson's disease, these signals often become stuck in a slow, repetitive loop known as beta activity. While doctors have long known that this abnormal rhythm is linked to the disease, the precise way these signals behave in the moments before a person loses the ability to walk has remained a mystery. This uncertainty is critical because one of the most disabling symptoms of Parkinson's is freezing of gait, a sudden, terrifying inability to take a step despite the intention to move. Understanding the exact nature of the brain's electrical activity right before this freeze occurs could offer a window into how the brain transitions from smooth walking to a complete standstill.

A recent study set out to examine this transition by listening directly to the electrical signals inside the brains of four individuals with Parkinson's disease. The researchers focused specifically on the timing of the bursts of beta activity in the subthalamic nucleus while the patients were walking. Instead of just looking at how strong these signals were, the team analyzed the irregularity and unpredictability of when these bursts happened. They compared the brain activity recorded during stable, normal walking with the activity recorded just before a freezing episode occurred. By using a method that tested how well these patterns could predict a freeze in one patient based on data from the others, the team sought to determine if the brain's rhythm changes in a specific, measurable way before the legs stop moving.

The investigation revealed a distinct shift in the brain's electrical behavior. Just before a freezing episode, the timing of the beta bursts became significantly more predictable and less variable. In other words, the brain's signals, which usually fluctuate with a certain amount of natural randomness, began to fall into a rigid, constrained pattern. This reduction in the unpredictability of the signal timing was a strong indicator that a freeze was about to happen. The researchers found that this change in timing could predict the onset of freezing with high accuracy, correctly identifying the event in the vast majority of cases. Once the freezing actually began, the brain's signals showed even less variation, settling into a highly structured state that was different from the stable walking state.

While the study also looked at how different frequencies of brain waves interacted with one another, the results there were mixed and varied from person to person. The primary discovery, however, was clear: the brain loses its temporal flexibility right before a person freezes. The findings suggest that freezing of gait is not merely a random failure of movement but a transition into a state where the brain's neural dynamics become overly constrained. This insight offers a concrete explanation for why the ability to walk can suddenly vanish, pointing to a specific change in the timing of electrical signals that precedes the physical stop. The study confirms that by monitoring the regularity of these bursts, it is possible to foresee the moment a person is about to freeze, providing a new way to understand the mechanics of this debilitating symptom.

Technical Summary: Reduced Entropy of Subthalamic Beta Bursts Predicts Freezing of Gait in Parkinson's Disease

Problem Statement
Freezing of gait (FOG) is a debilitating symptom in Parkinson's disease (PD) characterized by a sudden, transient inability to initiate or continue walking. While FOG is known to be associated with abnormal beta-band oscillatory activity within the subthalamic nucleus (STN), the specific temporal structure of these beta burst dynamics remains poorly understood. Existing literature has not sufficiently characterized how the timing and variability of these bursts differ between stable gait and the moments immediately preceding a freeze episode.

Methodology
The study analyzed local field potential (LFP) recordings from the STN of four individuals with Parkinson's disease during gait tasks. The researchers focused on extracting temporal features of beta bursts, specifically examining the timing of inter-burst intervals. Key metrics computed included entropy (a measure of unpredictability or disorder) and variability across different behavioral states: pre-freeze, stable gait, and freezing episodes. To evaluate the clinical utility of these features, the authors employed a leave-one-patient-out cross-validation approach to train classifiers capable of distinguishing between pre-freeze and stable gait states. Additionally, phase-amplitude coupling (PAC) was assessed to explore frequency-specific interactions, though these effects were noted to be heterogeneous.

Key Contributions and Results
The primary contribution of this work is the identification of reduced temporal entropy in STN beta burst intervals as a specific biomarker for impending freezing. The results demonstrated that:

Pre-freeze State: Entropy of inter-burst intervals was significantly reduced prior to the onset of freezing ( $p < 0.01$ ). This reduction indicates a transition toward more constrained and predictable neural dynamics before the freeze occurs.
Predictive Performance: Classifiers utilizing these temporal features achieved strong predictive performance, with an Area Under the Curve (AUC) of 0.825. When a specific threshold was applied, the AUC improved to 0.858.
Freezing State: During the actual freezing episode, variability measures decreased further, and temporal structure increased, although entropy levels did not differ significantly from the pre-freeze state.
Phase-Amplitude Coupling: While frequency-specific PAC effects were observed, they were heterogeneous across the comparisons, suggesting that burst timing entropy is a more robust predictor than PAC in this context.

Significance and Claims
The paper concludes that the reduced temporal variability of STN beta burst timing is a precursor to freezing of gait. The authors posit that this reduction in entropy reflects a transition to "constrained neural dynamics" within the STN, which may serve as a mechanistic driver for the loss of motor flexibility observed in FOG. The study claims that these temporal features are not only distinct markers of the pre-freeze state but also possess sufficient predictive power to potentially identify freezing onset before it manifests behaviorally. The findings suggest that monitoring the entropy of beta burst timing could offer a viable target for closed-loop deep brain stimulation strategies aimed at preventing freezing episodes.

Technical Summary: Reduced Entropy of Subthalamic Beta Bursts Predicts Freezing of Gait in Parkinson's Disease

More like this