Abstract
INTRODUCTION
Early identification of amyloid-beta (Aβ) pathology is essential for timely intervention in Alzheimer’s disease (AD), as Aβ accumulation precedes clinical symptoms by years.
METHODS
We predicted positron emission tomography (PET) -defined Aβ status in cognitively normal individuals using non-invasive resting-state dry electroencephalography (EEG). Brain effective and functional connectivity were quantified to characterize dynamic information transfer between regions. Features derived from dynamic brain connectivity were used to train machine learning models for Aβ classification.
RESULTS
EEG-based models can predict amyloid positivity with accuracy comparable to, or in some cases exceeding, that of plasma p-tau217 and previously reported predictors based on cognitive measures, demographics, and health factors, depending on the model.
DISCUSSION
Beyond near real-time Aβ inference, the framework identifies the most affected brain regions and connections associated with neurodegeneration, offering complementary information to blood-based biomarkers and supporting its potential as an affordable tool for early AD detection in primary care and population screening.
If you do not see content above, kindly GO TO SOURCE.
Not all publishers encode content in a way that enables republishing at Neuro.vip.
This post is Copyright: | October 29, 2026
Neuro-Dementia