Matching sparse sets of cardiac image cross-sections using large deformation diffeomorphic metric mapping algorithm

Authors:
Siamak Ardekani;Aastha Jain;Saurabh Jain;Theodore P. Abraham;Maria R. Abraham;Stefan Zimmerman;Raimond L. Winslow;Michael I. Miller;Laurent Younes
Affiliations:
Center for Imaging Sciences, Johns Hopkins University;Center for Imaging Sciences, Johns Hopkins University;Center for Imaging Sciences, Johns Hopkins University;Medical Institutions, Johns Hopkins University;Medical Institutions, Johns Hopkins University;Medical Institutions, Johns Hopkins University;Institute for Computational Medicine, Johns Hopkins University;Center for Imaging Sciences, Johns Hopkins University;Center for Imaging Sciences, Johns Hopkins University
Venue:
STACOM'11 Proceedings of the Second international conference on Statistical Atlases and Computational Models of the Heart: imaging and modelling challenges
Year:
2011

Citing 3
Cited 0

Large Deformation Diffeomorphic Metric Curve Mapping

International Journal of Computer Vision
Point Set Registration: Coherent Point Drift

IEEE Transactions on Pattern Analysis and Machine Intelligence
Surface matching via currents

IPMI'05 Proceedings of the 19th international conference on Information Processing in Medical Imaging

Quantified Score

Hi-index	0.00

Visualization

Abstract

The purpose of this study is to illustrate the application of large deformation diffeomorphic metric mapping to perform registration among sparsely sampled cardiac magnetic resonance imaging (MRI) data. To evaluate the performance of this method, we use two sets of data: 1) contours that are generated from sparsely sampled left ventricular sections and extracted from short axis cardiac MRI of patients with hypertrophic cardiomyopathy and 2) left ventricular surface mesh that is generated from higher resolution cardiac computed tomography image. We present two different discrepancy criteria, one based on a measure that is embedded in the dual of a reproducing kernel Hilbert space of functions for curves and the other is based on a geometric soft matching distance between a surface and a curve.