Share Email Print
cover

Proceedings Paper

Accurate registration of coronary arteries for volumetric CT digital subtraction angiography
Author(s): Marco Razeto; James Matthews; Saad Masood; Jill Steel; Kazumasa Arakita
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In the diagnosis of coronary artery disease with coronary computed tomography angiography, accurate evaluation remains challenging in the presence of calcifications or stents. Volumetric CT Digital Subtraction Angiography is a novel method that may become a powerful tool to overcome these limitations. However, precise registration of structures is essential, as even small misalignments can produce striking and disruptive bright and dark artefacts. Moreover, for clinical acceptance, the tool should require minimal user interaction and fast turnaround, thereby raising several challenges. In this paper we address the problem with a registration method based on a global non- rigid step, followed by local rigid refinement. In our quantitative analysis based on 10 datasets, each consisting of a pair of pre- and post-contrast scans of the same patient, we achieve an average Target Registration Error of 0.45 mm. Runtimes are less than 90 seconds for the global step, while each local refinement takes less than 15 seconds to run. Initial clinical evaluation shows good results in cases of moderate calcification, and indicates that around 50% of severely calcified and previously non-interpretable cases have been made interpretable by application of our method.

Paper Details

Date Published: 14 March 2013
PDF: 6 pages
Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 876834 (14 March 2013); doi: 10.1117/12.2010931
Show Author Affiliations
Marco Razeto, Toshiba Medical Visualization Systems (United Kingdom)
James Matthews, Toshiba Medical Visualization Systems (United Kingdom)
Saad Masood, Toshiba Medical Visualization Systems (United Kingdom)
Jill Steel, Toshiba Medical Visualization Systems (United Kingdom)
Kazumasa Arakita, Toshiba Medical Systems Corp. (Japan)


Published in SPIE Proceedings Vol. 8768:
International Conference on Graphic and Image Processing (ICGIP 2012)
Zeng Zhu, Editor(s)

© SPIE. Terms of Use
Back to Top