Individual cases of Parkinson’s disease can be robustly classified by cortical oscillatory activity from magnetoencephalography
Bibliographic record
Abstract
Abstract Parkinson’s disease (PD) is a progressive neurodegenerative disorder which causes debilitating symptoms in both the motor and cognitive domains. The neurophysiological markers of PD include ‘oscillopathies’ such as diffuse neural oscillatory slowing, dysregulated beta band activity, and changes in interhemispheric functional connectivity; however, the relative importance of these markers as determinants of disease status is not clear. In this study, we used resting state magnetoencephalography data (n = 199 participants, 78 PD, 121 controls) from the open OMEGA repository to investigate changes in spectral power and functional networks in PD. Using a Contrast of Parameter Estimates (COPE) approach, we modelled the effects of PD while controlling for population-level confounds (age, sex, brain volume). Permutation testing revealed highly significant increases in theta (p=0.0001) and decreases in gamma band spectral power (p=0.0001). Building on the group contrast results, we investigated the ability of source-resolved MEG data to distinguish PD from healthy controls. Our approach uses a Partial Least Squares (PLS)-based classifier to find linear combinations of MEG features which independently predict PD. We found MEG-based predictions to be highly sensitive and specific, reaching an optimal AUC-ROC of 0.87 ± 0.04 using a model including spectral power features with 4 independent PLS components, compared to 0.68 ± 0.04 when using functional connectivity. Interpretation of the model weights suggests that oscillatory slowing can be separated into independent posterior theta and global diffuse delta components that can robustly identify individual cases of PD with a high degree of accuracy. This suggests MEG can reveal dissociable, complementary neural processes which contribute to PD.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.011 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".