Share Email Print
cover

Proceedings Paper

Myocardial strain estimation from CT: towards computer-aided diagnosis on infarction identification
Author(s): Ken C. L. Wong; Michael Tee; Marcus Chen; David A. Bluemke; Ronald M. Summers; Jianhua Yao
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Regional myocardial strains have the potential for early quantification and detection of cardiac dysfunctions. Although image modalities such as tagged and strain-encoded MRI can provide motion information of the myocardium, they are uncommon in clinical routine. In contrary, cardiac CT images are usually available, but they only provide motion information at salient features such as the cardiac boundaries. To estimate myocardial strains from a CT image sequence, we adopted a cardiac biomechanical model with hyperelastic material properties to relate the motion on the cardiac boundaries to the myocardial deformation. The frame-to-frame displacements of the cardiac boundaries are obtained using B-spline deformable image registration based on mutual information, which are enforced as boundary conditions to the biomechanical model. The system equation is solved by the finite element method to provide the dense displacement field of the myocardium, and the regional values of the three principal strains and the six strains in cylindrical coordinates are computed in terms of the American Heart Association nomenclature. To study the potential of the estimated regional strains on identifying myocardial infarction, experiments were performed on cardiac CT image sequences of ten canines with artificially induced myocardial infarctions. The leave-one-subject-out cross validations show that, by using the optimal strain magnitude thresholds computed from ROC curves, the radial strain and the first principal strain have the best performance.

Paper Details

Date Published: 20 March 2015
PDF: 6 pages
Proc. SPIE 9414, Medical Imaging 2015: Computer-Aided Diagnosis, 941434 (20 March 2015); doi: 10.1117/12.2081464
Show Author Affiliations
Ken C. L. Wong, National Institutes of Health (United States)
Michael Tee, National Institutes of Health (United States)
Univ. of Oxford (United Kingdom)
Marcus Chen, National Institutes of Health (United States)
David A. Bluemke, National Institutes of Health (United States)
Ronald M. Summers, National Institutes of Health (United States)
Jianhua Yao, National Institutes of Health (United States)


Published in SPIE Proceedings Vol. 9414:
Medical Imaging 2015: Computer-Aided Diagnosis
Lubomir M. Hadjiiski; Georgia D. Tourassi, Editor(s)

© SPIE. Terms of Use
Back to Top