Estimation of positron-emission-tomography amyloid load and related biomarkers in Alzheimer’s disease using evoked potential tomography electroencephalography

Abstract

INTRODUCTION: Dementia affects over 50 million individuals globally, predominantly due to Alzheimer's disease (AD). Effective early detection, monitoring, and intervention remain clinical challenges, and existing diagnostic frameworks lack unified portable solutions for assessing multiple biomarkers. 

METHODS: This study evaluates Evoked Potential Tomography (EPT), an EEG-based technique using a novel visual evoked potential protocol. We developed an automated pipeline including EEG preprocessing, event-related potential feature extraction, feature selection, optimization, and regression modeling to estimate key AD biomarkers: PET-amyloid standardized uptake value ratio (SUVR), CSF phosphorylated tau 181 (p-tau181), Free and Cued Selective Reminding Test (FCSRT) scores, and Mini-Mental State Examination (MMSE) scores. 

RESULTS: Regression modeling of ERP features from early-AD participants yielded strong Spearman’s correlations between predicted and true values for PET-amyloid SUVR, p-tau181, FCSRT, and MMSE scores, with r = 0.8-0.94. 

DISCUSSION: Despite limitations, findings support EPT's potential for sensitive, accurate, and noninvasive early AD detection and progression monitoring. Further validation is ongoing.

Authors

Rauchmann, B. S., Hamet, J., Lai, J., Kafi, H., Rexwinkle, J., Brendel, M., Franzmeier, N., Kurz, C., Pogarell, O., Levin, J., Höglinger, G., & Perneczky, R.