TY - GEN
T1 - Personalization of cardiac fiber orientations from image data using the unscented Kalman filter
AU - Nagler, Andreas
AU - Bertoglio, Cristóbal
AU - Gee, Michael
AU - Wall, Wolfgang
PY - 2013
Y1 - 2013
N2 - In this work, we propose to estimate rule-based myocardial fiber model (RBM) parameters from measured image data, with the goal of personalizing the fiber architecture for cardiac simulations. We first describe the RBM, which is based on a space-dependent angle distribution on the heart surface and then extended to the whole domain through an harmonic lifting of the fiber vectors. We then present a static Unscented Kalman Filter which we use for estimating the degrees of freedom of the fiber model. We illustrate the methodology using noisy synthetic data on a real heart geometry, as well as real DT-MRI-derived fiber data. We also show the impact of different fiber distributions on cardiac contraction simulations.
AB - In this work, we propose to estimate rule-based myocardial fiber model (RBM) parameters from measured image data, with the goal of personalizing the fiber architecture for cardiac simulations. We first describe the RBM, which is based on a space-dependent angle distribution on the heart surface and then extended to the whole domain through an harmonic lifting of the fiber vectors. We then present a static Unscented Kalman Filter which we use for estimating the degrees of freedom of the fiber model. We illustrate the methodology using noisy synthetic data on a real heart geometry, as well as real DT-MRI-derived fiber data. We also show the impact of different fiber distributions on cardiac contraction simulations.
UR - http://www.scopus.com/inward/record.url?scp=84879816962&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-38899-6_16
DO - 10.1007/978-3-642-38899-6_16
M3 - Conference contribution
AN - SCOPUS:84879816962
SN - 9783642388989
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 132
EP - 140
BT - Functional Imaging and Modeling of the Heart - 7th International Conference, FIMH 2013, Proceedings
T2 - 7th International Conference on Functional Imaging and Modeling of the Heart, FIMH 2013
Y2 - 20 June 2013 through 22 June 2013
ER -