Item

Abstract

Parkinson's disease motor symptoms are associated with excessive beta oscillations in the subthalamic nucleus (STN). However, multi-modal signals recorded during and after implantation of deep brain stimulation (DBS) electrode can provide many other, more fine-grained features. Various studies are currently searching for the features that can be useful in providing a meaningful relationship with various Parkinson’s symptoms. Finding them will allow for more precise diagnostics and treatment.

Yet, studies correlating quantitative measures based on electroencephalogram (EEG) with motor symptoms have yielded no clear relationship [3], inviting further exploration.

We use a large dataset to systematically investigate relationships between quantitative EEG, local field potential (LFP) in STN, and motor symptoms. The LFP and EEG data are measured in 30s blocks during surgery for the deep brain stimulation electrode placement on different electrode depths.

We analyze the scale-free behavior of the EEG and LFP signals and their potential implications for Parkinson’s disease motor symptoms. We use Detrended Fluctuation Analysis (DFA) to quantify the scaling of the signal, with the DFA exponent quantifying this scaling analogously to the Hurst exponent. In our case, we are particularly interested in the dynamics of envelope amplitude modulation of cortical EEG [see Fig. 1A-D] and how it relates to the concurrently recorded signal in STN. Previous studies have shown that cortical signals exhibit scale-free behavior in the alpha and beta bands [1]. We additionally demonstrate that the DFA exponent of the alpha band strongly correlates with the power of the alpha band [see Fig. 1E].

Furthermore, previous studies have shown a correlation between cortical long-range temporal correlations (LRTC) and imaginary coherence in STN [2]. We extend these results to the larger patients cohort and include an investigation of the relationship with symptoms.