Blood detection of autoimmune encephalitis based on laser-induced breakdown spectroscopy and Raman spectroscopy

Area of Science:

• Biomedical Spectroscopy
• Medical Diagnostics
• Analytical Chemistry

Background:

• Autoimmune encephalitis (AE) incidence is rising, complicating clinical diagnosis.
• Current diagnostic criteria for AE are complex and time-consuming.
• There is a need for rapid, accurate diagnostic methods for AE.

Purpose of the Study:

• To explore the feasibility of using combined laser-induced breakdown spectroscopy (LIBS) and Raman spectroscopy for identifying autoimmune encephalitis (AE) in blood samples.
• To analyze the underlying mechanisms of AE diagnosis from atomic and molecular perspectives.
• To develop an improved diagnostic approach for AE.

Main Methods:

• Serum samples from healthy individuals and AE patients were analyzed using LIBS and Raman spectroscopy.
• Etched mesh silicon wafers were employed as substrates to minimize spectral variability.
• Fusion spectra were created by combining LIBS and Raman data, followed by feature selection using mutual information (MI).
• A long short-term memory (LSTM) model was utilized for classification.

Main Results:

• A total of 1785 LIBS and 1785 Raman spectra were collected from 119 participants.
• The fused LIBS-Raman spectra, with 537 selected features, achieved a 95.04% accuracy in AE diagnosis using an LSTM model.
• This fused spectral approach significantly improved diagnostic accuracy and reduced test time compared to individual techniques.
• Correlations between spectral features (e.g., Na, K, tryptophan) and cytokines were identified.

Conclusions:

• The fusion of LIBS and Raman spectroscopy presents a feasible and accurate method for identifying AE serum.
• This technique offers a rapid and minimally invasive approach to assist in the clinical diagnosis of AE.
• The study provides insights into the spectral biomarkers associated with AE, aiding in understanding disease mechanisms.