6 June 2000 Automated 3D-stent localization from intravascular ultrasound image sequences
Author Affiliations +
This paper presents a new method for automated detection of stents. The method consists of three sequential steps. At first, a 3D to 2D projection procedure is applied to find the global stent location. A local search strategy was designed, using a combination of characteristic image features to extract the sampling point candidates of the stent in each of the original cross-sectional images. Finally, the preselected points are accepted or rejected depending on a set of a priori criteria for position and shape. Using the resulting stent points, geometrical parameters of the stent are automatically calculated and a wire frame model is generated for 3D surface reconstruction. Thus, in combination with our algorithms for automated detection of the lumen cross-sectional area, the new method is an essential component for 3D visualization of stents and the automated quantification of the degree of in- stent restenosis. The evaluation is based on in vitro and in vivo recordings. The results show that the new algorithm is well-suited to replace time consuming manual segmentation and measurements.
© (2000) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michael Schmauder, Steffen Zeiler, C. M. Gross, Juergen Waigand, Reinhold Orglmeister, "Automated 3D-stent localization from intravascular ultrasound image sequences", Proc. SPIE 3979, Medical Imaging 2000: Image Processing, (6 June 2000); doi: 10.1117/12.387656; https://doi.org/10.1117/12.387656

Back to Top