In this section
@article{Yang:2013,
author = {Yang, GUANG and Hipwell, J and Hawkes, D and Arridge, S},
journal = {Annals of the British Machine Vision Association},
pages = {1--38},
title = {Numerical methods for coupled reconstruction and registration in digital breast tomosynthesis.},
volume = {2013},
year = {2013}
}
TY - JOUR
AB - Digital Breast Tomosynthesis (DBT) provides an insight into the fine details of normalfibroglandular tissues and abnormal lesions by reconstructing a pseudo-3D image of thebreast. In this respect, DBT overcomes a major limitation of conventional X-ray mam-mography by reducing the confounding effects caused by the superposition of breasttissue. In a breast cancer screening or diagnostic context, a radiologist is interested indetecting change, which might be indicative of malignant disease. To help automatethis task image registration is required to establish spatial correspondence between timepoints. Typically, images, such as MRI or CT, are first reconstructed and then registered.This approach can be effective if reconstructing using a complete set of data. However,for ill-posed, limited-angle problems such as DBT, estimating the deformation is com-plicated by the significant artefacts associated with the reconstruction, leading to severeinaccuracies in the registration.This paper presents a mathematical framework, which couples the two tasks andjointly estimates both image intensities and the parameters of a transformation. Underthis framework, we compare an iterative method and a simultaneous method, both ofwhich tackle the problem of comparing DBT data by combining reconstruction of a pairof temporal volumes with their registration.We evaluate our methods using various computational digital phantoms, uncom-pressed breast MR images, and in-vivo DBT simulations. Firstly, we compare both iter-ative and simultaneous methods to the conventional, sequential method using an affinetransformation model. We show that jointly estimating image intensities and parametrictransformations gives superior results with respect to reconstruction fidelity and regis-tration accuracy. Also, we incorporate a non-rigid B-spline transformation model intoour simultaneous method. The results demonstrate a visually plausible recovery of thedeformation w
AU - Yang,GUANG
AU - Hipwell,J
AU - Hawkes,D
AU - Arridge,S
EP - 38
PY - 2013///
SP - 1
TI - Numerical methods for coupled reconstruction and registration in digital breast tomosynthesis.
T2 - Annals of the British Machine Vision Association
VL - 2013
ER -
For enquiries about the MRI Physics Collective, please contact:
Mary Finnegan
Senior MR Physicist at the Imperial College Healthcare NHS Trust
Pete Lally
Assistant Professor in Magnetic Resonance (MR) Physics at Imperial College
Jan Sedlacik
MR Physicist at the Robert Steiner MR Unit, Hammersmith Hospital Campus