Category: Parkinson's Disease: Neurophysiology
Objective: To identify bursts in wide frequency bands and show the relation between burst features and clinical scores.
Background: Beta burst activity is a pathophysiological marker of Parkinson’s disease (PD). However, current methods to identify beta bursts focus on a single frequency bin with maximum power, inherently ignoring the variation of bursts across wider beta band activity. Building on existing methods, we develop a method for beta burst identification in a wide band.
Method: Seven subjects with advanced PD were recorded in resting state with Medtronic SenSight electrodes three (3) months after lead implantation in medication off. LFP signals contaminated with ECG artifacts were cleaned and decomposed using wavelet transform. Bursts as 2D regions were determined in time – frequency spectrum that exceed a threshold of 80%. Bursts thus straddle multiple frequency bins and stretch across time. To define a burst, we search for connected components in the time-frequency matrix, and for each burst, we identify the bounding region. Lastly, for each bounding region, we define four burst features: (i) duration, (ii) amplitude, (iii) Δf (span in the y-axis), and (iv) frequency.
Results: 1. Low beta band presents itself with bursts of longer duration bursts that is positively correlated with UPDRS III scores in medication off-state whereas the contrary is true for high beta band.
2. Bursts in low beta band have a higher probability of occurring, has higher bilateral synchronization across hemispheres, and have higher durations when compared to bursts high beta band.
Conclusion: We present a novel data-driven beta burst detection technique that can locate bursts in a wide frequency band. We demonstrate burst features that correlate with clinical scores, longer bursts in low beta correlating positively whereas longer bursts in high beta band correlating negatively. Potential findings point towards a two-band hypothesis, low beta that is more recruited, and with longer bursts linked to parkinsonism whereas high beta that could be linked to the status quo in cortico-basal ganglia loop.
To cite this abstract in AMA style:T. Sil, M. Muthuraman, H. Eldebakey, I. Hanafi, J. Volkmann, M. Reich, R. Peach. Wavelet-based bracketing, time-frequency beta-burst detection: new insights in Parkinson’s disease [abstract]. Mov Disord. 2023; 38 (suppl 1). https://www.mdsabstracts.org/abstract/wavelet-based-bracketing-time-frequency-beta-burst-detection-new-insights-in-parkinsons-disease/. Accessed September 22, 2023.
« Back to 2023 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/wavelet-based-bracketing-time-frequency-beta-burst-detection-new-insights-in-parkinsons-disease/