Spectral features of resting-state EEG in Parkinson's Disease: A multicenter study using functional data analysis

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Parkinson's Disease (PD) diagnosis relies on clinical symptoms, necessitating objective biomarkers.
• Resting-state electroencephalogram (rs-EEG) offers a non-invasive method to investigate brain activity.
• Spectral features of rs-EEG may reveal distinct patterns in PD.

Purpose of the Study:

• To analyze differences in rs-EEG spectral features between PD and healthy subjects using Functional Data Analysis (FDA).
• To validate and assess the reproducibility of these findings across four independent cohorts.
• To compare epoch-to-epoch FDA with averaged-epochs approaches for analyzing rs-EEG data.

Main Methods:

• Included 169 subjects (85 non-PD, 84 PD) from four centers.
• Extracted sensor-level relative power spectral density (PSD), dominant frequency (DF), and DF variability (DFV) from preprocessed rs-EEG signals.
• Compared PD and non-PD groups using both averaged epochs and epoch-to-epoch FDA.

Main Results:

• Significantly higher theta relative PSD in PD across all datasets for averaged epochs.
• Higher pre-alpha relative PSD observed in PD patients in three of four datasets.
• FDA revealed consistent posterior pre-alpha differences across multiple epochs in PD.

Conclusions:

• Increased generalized theta and posterior pre-alpha relative PSD are reproducible findings in PD.
• Rs-EEG spectral features, particularly theta and pre-alpha, demonstrate generalizability in PD.
• FDA is a reliable and powerful tool for analyzing epoch-to-epoch rs-EEG variations.