Ultrasensitive Detection of Blood-Based Alzheimer’s Disease Biomarkers: A Comprehensive SERS-Immunoassay Platform Enhanced by Machine Learning

• Published in: ACS chemical neuroscience 2024-12, Vol.15 (24), p.4390-4401
• Main Authors: Resmi, A. N., Nazeer, Shaiju S., Dhushyandhun, M. E., Paul, Willi, Chacko, Binu P., Menon, Ramshekhar N., Jayasree, Ramapurath. S.
• Format: Article
• Language: English
• Subjects: Aged; Alzheimer Disease - blood; Alzheimer Disease - diagnosis; Amyloid beta-Peptides - analysis; Amyloid beta-Peptides - blood; Biomarkers - blood; Cognitive Dysfunction - blood; Cognitive Dysfunction - diagnosis; Female; Humans; Immunoassay - methods; Machine Learning; Male; Spectrum Analysis, Raman - methods; tau Proteins - analysis; tau Proteins - blood
• ISSN: 1948-7193; 1948-7193
• DOI: 10.1021/acschemneuro.4c00369

Accurate and early disease detection is crucial for improving patient care, but traditional diagnostic methods often fail to identify diseases in their early stages, leading to delayed treatment outcomes. Early diagnosis using blood derivatives as a source for biomarkers is particularly important for managing Alzheimer’s disease (AD). This study introduces a novel approach for the precise and ultrasensitive detection of multiple core AD biomarkers (Aβ40, Aβ42, p-tau, and t-tau) using surface-enhanced Raman spectroscopy (SERS) combined with machine-learning algorithms. Our method employs an antibody-immobilized aluminum SERS substrate, which offers high precision, sensitivity, and accuracy. The platform achieves an impressive detection limit in the attomolar (aM) range and spans a wide dynamic range from aM to micromolar (μM) concentrations. This ultrasensitive and specific SERS immunoassay platform shows promise for identifying mild cognitive impairment (MCI), a potential precursor to AD, from blood plasma. Machine-learning algorithms applied to the spectral data enhance the differentiation of MCI from AD and healthy controls, yielding excellent sensitivity and specificity. Our integrated SERS-machine-learning approach, with its interpretability, advances AD research and underscores the effectiveness of a cost-efficient, easy-to-prepare Al-SERS substrate for clinical AD detection.