Investigations of atherosclerotic coronary arteries by confocal raman microspectroscopy

• Main Author: Chang, Henry C. Y.
• Language: English
• Published: 2009
• Online Access: http://hdl.handle.net/2429/13930

Atherosclerosis is a vascular disease, initiated by damage to the endothelial cells, and characterized by the deposition of plaque, which composes of mainly lipids and cholesterol, in the inner arterial wall that leads to progressive narrowing and hardening of the artery. The assessment of plaque status for the prediction of disease progression requires the examination of plaque's composition and structure rather than its size or gross morphology. Recent studies have shown Raman spectroscopy to be well suited for diagnosis of tissues because of its ability to identify and quantify biochemical molecules and structural motifs within intact cells. The ultimate utility of Raman technique may lie in its potential for the real-time assessment of atherosclerosis in vivo using fiber optic probes. This project employed confocal Raman microspectroscopy for the study of normal and diseased human coronary artery specimens in situ. A new Raman microscopy system for tissue spectroscopy has been set-up and characterized. Raman spectra were collected at various locations of the arterial wall, either along the radius covering its different layers, or concentrically outward from the lumen. Numerous algorithms were developed and tested for baseline removal, normalization and smoothing of Raman spectra obtained. Key analysis involving component analysis and generalized two-dimensional (2D) correlation followed. A model for chemical component analysis of eighteen major biochemicals found in the tissue/plaque was developed based on the application of singular value decomposition (SVD). Quantitative information on the relative amount of individual biomolecule was provided by the component analysis in which each Raman artery spectrum was modeled to a set of eighteen biochemical basis spectra. Generalized 2D Raman correlation spectroscopy was employed to identify and classify variations in spectral features occurring at different locations of the artery transverse section. The 2D correlation results have demonstrated the ability of 2D correlation to detect subtle spectral changes happening within a tiny area of the arterial wall, as well as to provide qualitative information about the order of appearance of individual constituent that complemented the chemical composition results. Possible future improvements and continuation of the project were proposed in the final chapter of this thesis as well.