Correlation of Longitudinal Intravascular Ultrasound Data for the Clinical Assessment of Coronary Artery Disease Progression

Virtual histology-intravascular ultrasound (VH-IVUS) has gained increasing utility in the cardiac catheterization laboratory, not only in determining underlying atherosclerotic lesion composition prior to stent placement, but also in clinical studies assessing the natural history of coronary artery disease (CAD) [1]. Furthermore, VH-IVUS has provided an excellent means of quantifying disease progression by comparing data sets collected over time (i.e., longitudinal studies) and potentially identifying rapidly progressing and potentially vulnerable plaques. One difficulty, however, in analyzing VH-IVUS derived CAD progression is the accurate co-registration of image sets collected over a period of time. Commonly, an expert VH-IVUS image reader reviews these image sets side-by-side on a display and co-registers images along the vessel main axis, herein axially co-registered, by identifying image locations relative to fiduciary anatomical markers (e.g., branches). Despite this method being the standard for analyzing CAD progression, it is limited by the inability to accurately co-register VH-IVUS data in the circumferential direction (i.e., rotating images such that their cylindrical coordinate bases coincide; herein circumferentially co-registered). Thus, a significant amount of information on focal plaque progression is lost that could provide a greater understanding of the natural evolution of CAD, the effects of various pharmaceutical agents (e.g., statins) on lesion composition changes, and the impact of local mechanical factors that induce plaque progression/regression and transformation.