Optimal fluorescence imaging of atherosclerotic human tissue

摘要

Multispectral imaging was investigated as a technique to exploit the spatial and spectral information available in the diagnosis of atherosclerosis. A diagnostic system is proposed that could provide direct viewing of a standard endoscopic or a total-fluorescence image. In addition, multiple spectral-feature images, each associated with a separate, narrow spectral band, could be obtained and processed to produce an optimized contrast image. In this study, a 'white-light' image and a total-fluorescence image were obtained. In addition, a three- dimensional, multispectral data set was generated, and two methods of utilizing this data were explored: (1) a per-pixel ratio of fluorescence intensities, and (2) an optimized superposition of the spectral-feature images. Fluorescence imaging is found to provide a rich data set possessing great potential for improving the detection and characterization of atheromatous disease.