The abstract is read at the publisher; the summary is JClub's.
In 455 patients with clinical EEG and FDG-PET within one year, four oscillation-linked EEG latent factors differ across FDG-PET phenotypes, with strongest slowing/rhythm shifts in dementia with Lewy bodies and dysexecutive AD.
- Why it matters: Neuroimaging can subtype neurodegenerative disorders, but low-cost EEG biomarkers are needed to expand access and improve phenotype-specific identification beyond general severity.
- What they did: The study decomposed clinical EEGs into six nonnegative tensor latent factors, compared factor expression across 14 StateViewer FDG-PET diagnostic classes, regressed factors on a subject-level slowing index, and linked factors to regional FDG uptake and cognition (n = 98).
- The result: Posterior theta-slow alpha became more phenotype-specific after slowing adjustment (eta-squared 0.082 to 0.147), while artifact-dominated factors showed no robust association; the slowing index tracked an FDG-PET pattern highly associated with DLB (r = 0.51, P 0.8) and poorly separated frontotemporal dementia phenotypes tied to 4-repeat tau.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.