Abstract
The objectives of this study were to evaluate: (1) the feasibility of generating three- dimensional (3-D) ultrasound (US) volumetric images of arterial segments from intravascular (IV) US images by retaining full range of gray levels; (2) the feasibility of volumetric quantitation of various arterial wall pathology from the 3-D volume US images of arterial segments. IVUS provides morphologic details of arterial wall diseases. This is seen as variation in gray levels. However, when a 3-D US image is generated currently, the full range of gray levels is not utilized. This limits optimal assessment of arterial wall pathology
Sequential cross-sectional IVUS images from 11 arterial segments consisting of various pathology were obtained in vitro by calibrated withdrawal of an IVUS catheter. These images were digitized by an 8 bit digitizer to retain full 256 gray levels of bright ness. 3-D volume generation was carried out using "ANALYZE" software. After the IVUS imaging, arterial segments were sectioned transversely in a 0.3-0.4 mm cross section and stained with hematoxylin, eosin and elastin. Geometrical measurements and gross morphological changes of the arterial segments were noted and correlated with the corre sponding section of the image from the three-dimensional volume. Arterial wall (continued on next page) (Abstract continued) pathology, its extent and its effect on lumen geometry were easily appreciated in multiple tomographic sections of a 3-D volume image. Similarly, arterial wall pathology was easily quantitated from 3-D volume. The above assessments were only feasible by retaining full range of gray levels in the 3-D volume image. This study indicates that (1) it is feasible to generate a 3-D US volume image by retaining full range of gray levels from IVUS images, (2) retaining full range of gray levels allows optimal assessment of arterial wall pathology and its extent in 3-D volume, and (3) IVUS allows quantitation of arterial wall pathology, and thereby one can assess the effect of intervention.
Get full access to this article
View all access options for this article.