Study on glycoprotein terahertz time-domain spectroscopy based on composite multiscale entropy feature extraction...

Area of Science:

• Biophysics
• Medical Physics
• Spectroscopy

Background:

• Tumor genesis involves protein glycosylation, making glycoproteins potential tumor markers due to higher consumption by cancer cells.
• Terahertz (THz) spectroscopy offers a non-ionizing method for analyzing biological molecules.

Purpose of the Study:

• To investigate the terahertz time-domain attenuated total reflection (ATR) technique for analyzing glycoprotein solutions.
• To explore the relationship between glycoprotein concentration and THz absorption.
• To evaluate feature extraction methods for glycoprotein identification using THz spectroscopy.

Main Methods:

• Terahertz time-domain attenuated total reflection (ATR) technique was used to analyze glycoprotein solutions across a concentration range (0.2–50 mg/ml).
• Composite multiscale entropy (CMSE) and Principal Component Analysis (PCA) were employed for feature extraction.
• K-means clustering was utilized for data analysis.

Main Results:

• A significant nonlinear relationship was observed between the absorption coefficient and glycoprotein concentration.
• The CMSE method demonstrated superior specificity and sensitivity compared to PCA for glycoprotein recognition.
• Absorption coefficient and dielectric loss angle tangent were identified as suitable parameters for qualitative identification.

Conclusions:

• The study highlights the potential of THz spectroscopy and the CMSE method for the identification of glycoprotein-related tumor markers.
• The findings suggest THz technology could be valuable for medical applications in cancer diagnostics.
• Understanding the THz absorption dynamics, including THz excess and THz defect, is crucial for interpreting glycoprotein spectra.