Raman liquid biopsy: a new approach to the multiple sclerosis diagnostics.

Area of Science:

• Biomedical Spectroscopy
• Machine Learning in Diagnostics
• Neuroimmunology

Background:

• Multiple sclerosis (MS) lacks a reliable diagnostic biomarker.
• Current diagnostic methods for MS are often expensive or lack specificity.
• There is an urgent need for sensitive, specific, and accessible MS diagnostic tools.

Purpose of the Study:

• To investigate the utility of surface-enhanced Raman spectroscopy (SERS) of blood serum for MS diagnosis.
• To develop a machine learning model for differentiating MS patients from healthy controls using SERS data.

Main Methods:

• Blood serum samples were analyzed using surface-enhanced Raman spectroscopy (SERS).
• Projection on latent structures-discriminant analysis (PLS-DA) was employed for Raman spectra processing.
• Machine learning was used to classify individuals as having MS or being healthy.

Main Results:

• The SERS-based method achieved an average specificity of 0.96 and an average sensitivity of 0.89.
• Key Raman bands (632, 721-735, 1048-1076 cm⁻¹) were identified for discriminating between MS patients and healthy individuals.
• The study demonstrated the potential of SERS for MS diagnosis.

Conclusions:

• Blood serum SERS combined with machine learning shows promise as a diagnostic method for multiple sclerosis.
• This approach offers a potentially specific, sensitive, and accessible diagnostic strategy.
• Further comprehensive research is warranted to validate and refine this diagnostic technique.