Attenuation and backscattering based tissue characterization in intravascular optical coherence tomography pullback-runs (Conference Presentation)

Intravascular optical coherence tomography (IVOCT) is a new intravascular imaging modality which enables arterial structures to be visualized at a microstructure level. The determination of these structures is currently performed manually based on relative light intensities which is difficult because there are many factors, including the position inside the artery and vendor of the catheter, which can influence these intensities. In this study we demonstrate how optical attenuation and backscattering values can be computed and used as better characterizing features for different types of atherosclerotic plaque such as fibro-atheroma, lipid-pools and calcified areas. To validate the method, different plaque components are segmented in multiple IVOCT pullback runs using matching histology-data. The optical attenuation, backscattering and light intensity features of the segmented regions are then automatically extracted and analyzed for their entropy with regards to tissue characterization. The results of the validation analysis show that the computed attenuation and backscattering measurements are in agreement with those published in literature and that especially attenuation is a more robust feature than light intensity when it comes to tissue characterization. As a practical application we show how attenuation and backscattering can be used to quickly determine the presence of lipid or calcified plaques which can be important factors to determine patient treatment. Based on these findings we intend to develop a fully automatic tissue characterization method for IVOCT.