Unlocking Preclinical Alzheimer's: A Multi-Year Label-Free In Vitro Raman Spectroscopy Study Empowered by...

Area of Science:

• Biomedical Engineering
• Neuroscience
• Analytical Chemistry

Background:

• Alzheimer's disease (AD) is a progressive neurodegenerative disorder.
• Early detection is critical for intervention and clinical trial enrollment.
• Current diagnostic methods are often invasive, expensive, or require post-mortem analysis.

Purpose of the Study:

• To investigate label-free Raman spectroscopy and machine learning as a non-invasive tool for preclinical Alzheimer's diagnosis.
• To assess the robustness of the diagnostic model across different patient cohorts and measurement years.

Main Methods:

• Raman spectroscopy was used to analyze dried cerebrospinal fluid samples.
• Partial least squares discriminant analysis with variable selection identified key biomolecules.
• Machine learning models were developed for classification of preclinical Alzheimer's disease.

Main Results:

• Discriminative molecules including nucleic acids, amino acids, proteins, and carbohydrates (e.g., taurine/hypotaurine, guanine) were identified.
• A unified classification model demonstrated remarkable robustness across different cohorts and measurement years.
• The model successfully classified preclinical Alzheimer's disease, despite Raman spectroscopy's sensitivity to measurement variations.

Conclusions:

• Label-free Raman spectroscopy coupled with machine learning is a viable method for detecting preclinical Alzheimer's disease.
• This technique offers a promising, non-invasive alternative for early AD diagnosis.
• The findings open avenues for future clinical applications and improved diagnostic strategies.