Please use this identifier to cite or link to this item:
Title: An adaptive patient specific deformable registration for breast images of positron emission tomography and magnetic resonance imaging using finite element approach
Authors: Xue, C
Tang, FH 
Keywords: Breast
Magnetic resonance imaging
Positron emission tomography
Medical imaging
Finite element methods
Image registration
Issue Date: 2014
Publisher: SPIE-International Society for Optical Engineering
Source: Proceedings of SPIE : the International Society for Optical Engineering, 2014, v. 9034, 90343H How to cite?
Journal: Proceedings of SPIE : the International Society for Optical Engineering 
Abstract: A patient specific registration model based on finite element method was investigated in this study. Image registration of Positron Emission Tomography (PET) and Magnetic Resonance imaging (MRI) has been studied a lot. Surface-based registration is extensively applied in medical imaging. We develop and evaluate a registration method combine surface-based registration with biomechanical modeling. .Four sample cases of patients with PET and MRI breast scans performed within 30 days were collected from hospital. K-means clustering algorithm was used to segment images into two parts, which is fat tissue and neoplasm [2]. Instead of placing extrinsic landmarks on patients’ body which may be invasive, we proposed a new boundary condition to simulate breast deformation during two screening. Then a three dimensional model with meshes was built. Material properties were assigned to this model according to previous studies. The whole registration was based on a biomechanical finite element model, which could simulate deformation of breast under pressure.
Description: Medical Imaging 2014 : Image Processing, San Diego, CA, 15 Feb, 2014
ISSN: 0277-786X
EISSN: 1996-756X
DOI: 10.1117/12.2043252
Appears in Collections:Conference Paper

View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

Last Week
Last month
Citations as of Feb 19, 2019

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.