Disruption of low-frequency narrowband EEG microstate networks in Parkinson's disease with mild cognitive impairment.
Guangying Pei, Mengxuan Hu, Yiliu He and 7 others
PMID 41126224WHAT IT FOUND
Low-frequency brain-network patterns differed in people with Parkinson's and mild cognitive impairment, and these differences were associated with MoCA scores.
The all-frequency EEG signal did not show the same microstate parameter differences.
Key findings
01Narrowband EEG microstate parameters showed significant group differences in the delta, theta, and alpha frequency bands, while broadband microstate parameters showed no significant interaction for coverage, duration, or occurrence.
02In the delta band, MoCA scores were negatively correlated with coverage and occurrence of MS-A and with temporal and spatial variability of the whole-brain MS-A network.
03A machine-learning model using microstate features classified the three groups with mean accuracy 94.07% and classified PD-MCI versus PD-NC with mean accuracy 98.00%.
STILL TO COME
How it was doneWhat they foundWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study included only 67 people total, with 21 PD-MCI, 26 PD-NC, and 20 healthy controls, so the findings may not apply to other populations. It was cross-sectional, so it cannot show that EEG changes cause cognitive impairment or predict decline over time. People with Parkinson's were receiving treatment, and the study did not control exact medication timing or include off-drug recordings, so drug effects and disease effects are mixed. Cognitive status was classified mainly by MoCA scores, which the paper notes have sensitivity 41%, specificity 82%, and accuracy 68%, so some participants may be misclassified. The machine-learning accuracy was based on internal cross-validation in the same small sample, not external validation, so the 94.07% and 98.00% accuracies may not hold in routine clinical use.
The easy way to misread this
Do not read the 94.07% and 98.00% classification accuracies as evidence that EEG microstates can diagnose Parkinson's cognitive impairment in routine practice. They came from a small cross-sectional sample with internal cross-validation only, and the paper does not report external clinical testing.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →