Results 1 -
6 of
6
Optimization transfer approach to joint registration/reconstruction for motion-compensated image reconstruction," presented at the ISBI
, 2010
"... Motion artifacts in image reconstruction problems can be reduced by performing image motion estimation and image reconstruction jointly using a penalized-likelihood cost func-tion. However, updating the motion parameters by conven-tional gradient-based iterations can be computationally de-manding du ..."
Abstract
-
Cited by 5 (0 self)
- Add to MetaCart
(Show Context)
Motion artifacts in image reconstruction problems can be reduced by performing image motion estimation and image reconstruction jointly using a penalized-likelihood cost func-tion. However, updating the motion parameters by conven-tional gradient-based iterations can be computationally de-manding due to the system model required in inverse prob-lems. This paper describes an optimization transfer approach that leads to minimization steps for the motion parameters that have comparable complexity to those needed in image registration problems. This approach can simplify the imple-mentation of motion-compensated image reconstruction (MCIR) methods when the motion parameters are estimated jointly with the reconstructed image.
Motion compensated tomography reconstruction of coronary arteries in rotational angiography
- in IEEE Trans. Biomed. Eng
"... HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte p ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
(Show Context)
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et a ̀ la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
MOTION ASPECTS IN JOINT IMAGE RECONSTRUCTION AND NONRIGID MOTION ESTIMATION
, 2009
"... I have been enjoying my Ph.D. program in Ann Arbor a lot not only because of all the interesting research topics, but also because of all people I have in my life. Without them, I wouldn’t be able to survive and finish this program. I would like to thank God for sending me to Michigan and orchestrat ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
I have been enjoying my Ph.D. program in Ann Arbor a lot not only because of all the interesting research topics, but also because of all people I have in my life. Without them, I wouldn’t be able to survive and finish this program. I would like to thank God for sending me to Michigan and orchestrating my life among all people I met in Ann Arbor. The personal walk with Him was the source of my strength for the last 6 years. I thank my late father, Ye Ki Jun, for supporting me in many different ways. Without his promise to support me for a year, I wouldn’t be able to come to the United States. I know that it would be him to be the happiest person for my Ph.D., but I am sad that he couldn’t enjoy this moment with my family. Many people will miss him. However, I am still thankful that I can share my joy with my family, my mother Soon Ja Park, my sister Mi Hyun Jeon, my brother-in-law Masafumi Fuchikami, and my nephew Harumitsu Fuchikami. Family gatherings in Korea and Japan always cheered me up. I would like to thank my advisor Jeff Fessler for many things. I have been really enjoying working with him. His intuition for the problem, his passion for the research, and his openness to any possibilities such as my naive ideas have inspired me and I’ve learned a lot from him. It has been my privilege to work with him. I would also like to thank
Statistical Image Reconstruction and Motion Estimation for Image-Guided Radiotherapy
, 2011
"... To my parents, husband and daughter. ii ACKNOWLEDGEMENTS This thesis would have been impossible without my advisors, Prof. Jeff Fessler and Prof. James Balter. Working with them has been enjoyable and rewarding. They have understood difficulties from both my research and personal life, have always b ..."
Abstract
- Add to MetaCart
(Show Context)
To my parents, husband and daughter. ii ACKNOWLEDGEMENTS This thesis would have been impossible without my advisors, Prof. Jeff Fessler and Prof. James Balter. Working with them has been enjoyable and rewarding. They have understood difficulties from both my research and personal life, have always been supportive and willing to provide extra help when needed. I especially thank Jeff for his guidance, support and patience before I was formally enrolled at the University of Michigan. His deep insight and passion for research and dedication to his students are amazing. James always noticed my progress and encouraged me to accomplish more. I especially thank him for giving me freedom to pursue my research interest and easy access to clinical data, and setting our meetings at EECS for my convenience when I was pregnancy.
Communications Motion Compensated Tomography Reconstruction of Coronary Arteries in Rotational Angiography
, 2009
"... Abstract—This paper deals with the 3-D reconstruction of the coronary tree from a rotational X-ray projection sequence. It describes the following three stages: the reconstruction of the 3-D coronary tree at different phases of the cardiac cycle, the motion estimation, and the motion-compensated tom ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—This paper deals with the 3-D reconstruction of the coronary tree from a rotational X-ray projection sequence. It describes the following three stages: the reconstruction of the 3-D coronary tree at different phases of the cardiac cycle, the motion estimation, and the motion-compensated tomographic reconstruction of the 3-D coronary tree at one given phase using all the available projections. Our method is tested on a series of simulated images computed from the projection of a segmented dynamic volume sequence acquired in multislice computed tomography imaging. Performances are comparable to those obtained by reconstruction of a statical coronary tree using an algebraic reconstruction technique algorithm. Index Terms—Angiography, B-spline interpolation, deformable model, inverse problem, penalized least squares, rotational X-ray. I.
Abstract An Expanded Theoretical Treatment of Iteration-Dependent Majorize-Minimize Algorithms
, 2006
"... has received considerable attention in signal and image processing applications, as well as in the statistics literature. At each iteration of an MM algorithm, one constructs a tangent majorant function that majorizes the given cost function and is equal to it at the current iterate. The next iterat ..."
Abstract
- Add to MetaCart
(Show Context)
has received considerable attention in signal and image processing applications, as well as in the statistics literature. At each iteration of an MM algorithm, one constructs a tangent majorant function that majorizes the given cost function and is equal to it at the current iterate. The next iterate is obtained by minimizing this tangent majorant function, resulting in a sequence of iterates that reduces the cost function monotonically. A wellknown special case of MM methods are Expectation-Maximization (EM) algorithms. In this paper, we expand on previous analyses of MM, due to [12, 13], that allowed the tangent majorants to be constructed in iteration-dependent ways. Also, in [13], there was an error in one of the steps of the convergence proof that this paper overcomes. There are three main aspects in which our analysis builds upon previous work. Firstly, our treatment relaxes many assumptions related to the structure of the cost function, feasible set, and tangent majorants. For example, the cost function can be non-convex and the feasible set for the problem can be any convex set. Secondly, we propose convergence conditions, based on upper curvature bounds, that can be easier to verify than more standard continuity conditions. Furthermore, these conditions in some cases allow for considerable design freedom in the iteration-dependent behavior of the algorithm. Finally, we give an original characterization of the local region of convergence of MM algorithms based on connected (e.g., convex) tangent majorants. For such algorithms, cost function minimizers will locally attract the iterates over larger neighborhoods than is typically
