Comparative Study Into the Effect of Detector Noises and Sensitivity on the Serum SERS Analysis: Example of Non-Communicable Diseases Discrimination.

Abstract

The aim of the study is to compare the performance of surface-enhanced Raman spectroscopy (SERS) analysis of serum using a non-cooled detector (EnSpectr R785) and a high spectral resolution detector (Renishaw) in the task of discrimination between the patients with chronic heart failure and the patients with chronic obstructive pulmonary disease. The SERS-based solution of the classification problem demonstrates an insignificant relationship between the disease classification accuracy and detector quality (classification accuracy for the high-resolution setup is 0.84 and for the low-cost setup is 0.81). In the data recorded on Renishaw setup, the most significant bands are 611, 675, 720, 804, 1187, 1495, and 1847 cm^-1; for the EnSpectr setup-721, 1051, and 1665 cm^-1. The results have revealed equal discrimination capabilities in the EnSpectr and the Renishaw setups; however, the Renishaw setup has more prospects for identifying the spectral contribution of pathologically associated analytes.