Wirelessly transmitted subthalamic nucleus signals predict endogenous pain levels in Parkinson's disease patients

TL;DR

This study demonstrates that wirelessly transmitted subthalamic nucleus signals can predict pain fluctuations in Parkinson's disease patients, highlighting bilateral STN activity's role in endogenous pain regulation.

Contribution

It introduces a data-driven, machine learning approach to predict pain levels from STN signals, revealing bilateral STN biomarkers associated with pain fluctuations in PD.

Findings

Gamma and beta bands in contralateral STN influence pain prediction.

Delta and theta bands in ipsilateral STN are significant for pain prediction.

Bilateral STN activity impacts endogenous pain fluctuations.

Abstract

Parkinson disease (PD) patients experience pain fluctuations that significantly reduce their quality of life. Despite the vast knowledge of the subthalamic nucleus (STN) role in PD, the STN biomarkers for pain fluctuations and the relationship between bilateral subthalamic nucleus (STN) activities and pain occurrence are still less understood. This observational study used data-driven methods by collecting annotated pain followed by a series of corresponding binary pain ratings and wirelessly transmitted STN signals, then leveraging the explainable machine learning algorithm to predict binary pain levels and sort the feature influence. The binary pain levels could be predicted among annotated pain reports corresponding to PD-related pain characteristics. The STN activity from both sides could impact pain prediction, with gamma and beta bands in the contralateral STN and delta and theta bands in the ipsilateral STN showing a prominent role. This study emphasizes the role of bilateral STN biomarkers on endogenous pain fluctuations.