Abstract

This dissertation explores the applications of Raman spectroscopy for biological tissue analysis. Basics of Raman and surface-enhanced Raman spectroscopy (SERS) are described, along with an analysis of the literature on biodetection and non-invasive blood analysis with emphasis on glucose detection. Instrumentation for Raman measurements is described. The sources of noise and interferences occurring in biological tissue measurements are presented. The experimental work was focused on two distinct goals: a rapid detection of drugs in whole human blood with SERS; and reduction of the individual variability in non-invasive Raman measurements. A surface-enhanced Raman spectroscopy in NIR is utilized for a qualitative and quantitative analysis of drugs in blood. Optimization of this technique may become clinically viable for a rapid detection of drugs in blood samples. The challenge of non-invasive detection of molecules in blood through skin, without perforation, is the natural variability of skin optical properties and chemical composition. The proposed solution to this problem is to reduce the influence of the signal from tissue and distinguish the signal from the blood components by exploiting periodic changes of the blood content, causes by the blood pulse, while the tissues remain the same. A signal processing approach was developed where the natural blood pulse is identified from the series of Raman spectra, and used in a synchronous detection. The result of the algorithm is a Raman spectrum correlated with the Raman scattering on blood, while the uncorrelated signal from static tissues is diminished. The experimental proof for the two goals is presented and discussed. A series of appended publications form the basis of the dissertation.

Author (1)