The neurophysiological brain-fingerprint of Parkinson's disease

Abstract

In this study, we investigate the clinical potential of brain-fingerprints derived from electrophysiological brain activity for diagnostics and progression monitoring of Parkinson's disease (PD). We obtained brain-fingerprints from PD patients and age-matched healthy controls using short, task-free magnetoencephalographic recordings. The rhythmic components of the individual brain-fingerprint distinguished between patients and healthy participants with approximately 90% accuracy. The most prominent cortical features of the Parkinson's brain-fingerprint mapped to polyrhythmic activity in unimodal sensorimotor regions. Leveraging these features, we also show that Parkinson's disease stages can be decoded directly from cortical neurophysiological activity. Additionally, our study reveals that the cortical topography of the Parkinson's brain-fingerprint aligns with that of neurotransmitter systems affected by the disease's pathophysiology. We further demonstrate that the arrhythmic components of cortical activity are more variable over short periods of time in patients with Parkinson's disease than in healthy controls, making individual differentiation between patients based on these features more challenging and explaining previous negative published results. Overall, we outline patient-specific rhythmic brain signaling features that provide insights into both the neurophysiological signature and clinical staging of Parkinson's disease. For this reason, the proposed definition of a rhythmic brain-fingerprint of Parkinson's disease may contribute to novel, refined approaches to patient stratification and to the improved identification and testing of therapeutic neurostimulation targets.

Keywords: Movement disorders; Parkinson’s disease; arrhythmic brain activity; brain-fingerprinting; magnetoencephalography; neural dynamics; oscillations.

Figure 1:. Brain-fingerprinting pipeline and study design.

(a) From each participant, the power spectrum density of MEG source time series is computed for each region defined by the Desikan-Killiany atlas. This is done for two data segments (datasets 1 and 2), each containing approximately 4 minutes of clean data. The power spectra from these segments form two spectral brain-fingerprints (b-fp1 and b-fp2) for each participant . A confusion matrix, using self- and other-similarity measures of these brain-fingerprints across participants, enables inter-individual differentiation assessment. (b) We evaluated the effectiveness of this approach in differentiating individuals among three groups: i) healthy controls, ii) patients with Parkinson’s Disease (PD), and iii) each PD patient compared to healthy controls. (c) We derived an individual differentiability score for each participant, based on the self-similarity of their two brain-fingerprints. This score is z-scored against the other-similarity of their fingerprints with those of other participants in the study.

Figure 2:. Differentiating Patients with Parkinson’s Disease from Healthy Controls Using Spectral Brain-Fingerprints.

(a) Accuracy in distinguishing participants from their brain-fingerprints derived from full, arrhythmic, and rhythmic neurophysiological power spectra, estimated from 4-minute (bar plots) and 30-second (scatter plots) data segments. Scatter plots indicate differentiation accuracy for all brain-fingerprint pairs derived from all possible contiguous 30-second segments derived from the original 4-minute recordings. Grey segments at the base of the bar plots indicate control differentiation performances based on empty-room MEG recordings collected during each participant’ visit (refer to Methods). Error bars represent bootstrapped 95% confidence intervals. (b) Self-similarity statistics within participants for full spectral, rhythmic, and arrhythmic brain-fingerprints. The plots show the empirical density of self-similarity statistics between two consecutive brain-fingerprints in control and PD cohorts, with the PD group showing a wider distribution, suggesting more variability in patients for full spectral and aperiodic features. (c) Self-similarity of brain-fingerprints from brief (30-second) brain data segments across full spectral, rhythmic, and arrhythmic features. PD patients show lower self-similarity with increased gap durations between recordings (y-intercept shift downwards). The self-similarity of patient full spectrum brain-fingerprints decreases more rapidly as the gap duration between recordings increases. In contrast, the self-similarity of patient brain-fingerprints from rhythmic components was more self-similar than controls at short gap durations, and became comparable at longer durations. Shaded regions indicate the standard error on the mean.

Figure 3:. Comparative Analysis of Brain-Fingerprint Differentiation in Parkinson’s Disease and Control Groups

(a) Cortical maps comparing ICC scores for differentiating between patients and controls. Orange areas show regions where differentiation of individual patients is more effective than in controls. We replicated this finding in two independent samples of healthy controls: the Prevent-AD dataset (top panel) and the CamCAN dataset (bottom panel). (b) Differentiation accuracy from brain-fingerprints defined by top features for differentiating patients (left, cortical areas shown in orange) and top features for differentiating controls (right, cortical areas shown in blue).

Figure 4:. Decoding Stages of Parkinson’s Disease from Brain-Fingerprints.

(a) Cortical topography of decoding accuracies for Parkinson’s disease stages (based on binarized Hoehn & Yahr scores). On the right, power spectra of resting-state neurophysiological activity in the right postcentral gyrus, the cortical region with the highest accuracy in disease stage decoding. Plots represent the average power spectrum for each group: healthy controls, early and advanced disease stages, with shaded areas indicating standard errors across groups. (b) Scatter plot showing how the decoding accuracy of Parkinson’s disease stages from brain-fingerprint features of each cortical parcel correlates with the saliency of each parcel, as determined by its ΔICC score.

Figure 5:. Correlation of Spectral Brain-Fingerprints with Cortical Functional Hierarchy and Neurotransmitter Systems.

(a) Top: Cortical map illustrating the first unimodal-to-transmodal functional gradient, sourced from neuromaps. Bottom: Linear association between the weights of cortical regions in this functional gradient (as per neuromaps) and their prominence in the PD brain-fingerprint (Figure 3a, top). (b) Top: Bayes factor analysis of the topographical alignment between PD brain-fingerprint features (from Figure 3a) and atlases of various cortical neurochemical systems, highlighting strong correlations particularly with serotonin, cannabinoid, mu-opioid, and norepinephrine systems. Each row represents data from different control samples. Bottom: Selected neurochemical cortical atlases, as obtained from neuromaps.