Towards real-time 3D US to CT bone image registration using phase and curvature feature based GMM matching

Authors:
Anna Brounstein;Ilker Hacihaliloglu;Pierre Guy;Antony Hodgson;Rafeef Abugharbieh
Affiliations:
Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada;Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada;Department of Orthopaedics, University of British Columbia, Vancouver, Canada;Department of Mechanical Engineering, University of British Columbia, Vancouver, Canada;Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada
Venue:
MICCAI'11 Proceedings of the 14th international conference on Medical image computing and computer-assisted intervention - Volume Part I
Year:
2011

Citing 6
Cited 2

Modern Differential Geometry of Curves and Surfaces with Mathematica

Modern Differential Geometry of Curves and Surfaces with Mathematica
A Stochastic Iterative Closest Point Algorithm (stochastICP)

MICCAI '01 Proceedings of the 4th International Conference on Medical Image Computing and Computer-Assisted Intervention
Bone Segmentation and Fracture Detection in Ultrasound Using 3D Local Phase Features

MICCAI '08 Proceedings of the 11th international conference on Medical Image Computing and Computer-Assisted Intervention - Part I
Robust Point Set Registration Using Gaussian Mixture Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Self-calibrating ultrasound-to-CT bone registration

MICCAI'05 Proceedings of the 8th international conference on Medical Image Computing and Computer-Assisted Intervention - Volume Part I
Registration of CT and Intraoperative 3-D Ultrasound Images of the Spine Using Evolutionary and Gradient-Based Methods

IEEE Transactions on Evolutionary Computation

Non-iterative multi-modal partial view to full view image registration using local phase-based image projections

IPCAI'12 Proceedings of the Third international conference on Information Processing in Computer-Assisted Interventions
3D Ultrasound-CT registration in orthopaedic trauma using GMM registration with optimized particle simulation-based data reduction

MICCAI'12 Proceedings of the 15th international conference on Medical Image Computing and Computer-Assisted Intervention - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to use pre-operatively acquired computed tomography (CT) scans to guide surgical tool movements in orthopaedic surgery, the CT scan must first be registered to the patient's anatomy. Threedimensional (3D) ultrasound (US) could potentially be used for this purpose if the registration process could be made sufficiently automatic, fast and accurate, but existing methods have difficultiesmeeting one or more of these criteria.We propose a near-real-time US-to-CT registration method that matches point clouds extracted from local phase images with points selected in part on the basis of local curvature. The point clouds are represented as Gaussian Mixture Models (GMM) and registration is achieved by minimizing the statistical dissimilarity between the GMMs using an L2 distance metric. We present quantitative and qualitative results on both phantom and clinical pelvis data and show a mean registration time of 2.11 s with a mean accuracy of 0.49 mm.