東華大學圖書館 |

Tools for Coronary MR Angiography: Non-Cartesian Trajectories and Motion-Corrected Reconstructions.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Tools for Coronary MR Angiography: Non-Cartesian Trajectories and Motion-Corrected Reconstructions./
作者:	Kwang Eun Jang.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2024,
面頁冊數:	129 p.
附註:	Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
Contained By:	Dissertations Abstracts International85-11B.
標題:	Biomedical engineering. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31255811
ISBN:	9798382231112

Tools for Coronary MR Angiography: Non-Cartesian Trajectories and Motion-Corrected Reconstructions.
Kwang Eun Jang.

Tools for Coronary MR Angiography: Non-Cartesian Trajectories and Motion-Corrected Reconstructions. - Ann Arbor : ProQuest Dissertations & Theses, 2024 - 129 p.

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

Thesis (Ph.D.)--Stanford University, 2024.

X-ray coronary angiography remains the predominant technique for coronary artery examination, yet it exposes both patients and operators to ionizing radiation and is invasive. Despite strong demand for non-invasive and safer alternatives, the clinical adoption of coronary MR angiography has been slow. The primary challenges in coronary MR angiography are its prolonged scan time and limited resolution. Because coronary arteries often follow tortuous paths and have relatively small diameters, high resolving power is required. Additionally, whole-heart imaging is preferred over target-volume methods, as it reduces operator dependency. Achieving large volumetric coverage at high resolution demands collecting a significant amount of data. However, continuous myocardial movement restricts data collection to specific cardiac cycles, necessitating the segmented data acquisition over multiple heartbeats. Consequently, coronary MR angiography is slow and susceptible to both respiratory and cardiac motion. This dissertation introduces several tools for coronary MR angiography categorized into non-Cartesian trajectory design and non-rigid motion-corrected reconstruction. Non-Cartesian trajectories often exhibit higher sampling efficiency and greater robustness against undersampling compared to traditional rectilinear approaches, making them particularly useful for reducing the overall scan duration. Because non-rigid models capture a wide range of transformations, non-rigid motion-corrected reconstruction reduces resolution loss due to respiratory and cardiac motion.More specifically, this thesis introduces a unified methodology for designing three commonly used non-Cartesian trajectories: 3D Radial, 3D Cones, and Stack-of-spirals. The fundamental idea of the proposed method is that a non-Cartesian trajectory can be regarded as a discretized version of an analytic coordinate defined with a set of template trajectories. In the proposed method, discretizing the analytic coordinate is equivalent to identifying rotation angles for template trajectories, yielding the non-Cartesian trajectory. The discretization is achieved by creating a spiral path on a surface and sampling points along this path. This approach also allows for the analytical computation of the density compensation factor and the incorporation of useful features such as variable-density and spherical design.Additionally, this dissertation introduces image-space gridding for representing non-rigid motion. Similar to "gridding," a technique commonly employed to resample non-Cartesian k-space data onto a Cartesian grid, image-space gridding consists of an exact forward-ad joint pair of linear operators that resample an image onto a Cartesian grid warped by a displacement field. Image-space gridding allows incorporating nonrigid transformations into the data acquisition model. This formulation accounts for the inhomogeneity of data collected at different states of a target object; in the context of cardiac MR imaging, this formulation accounts for non-rigid transformations of the heart during the segmented data acquisition over multiple heartbeats.As a demonstration, this thesis presents non-rigid motion-corrected reconstructions for free-breathing coronary MR angiography. Following translational motion correction, 3D self-navigating image-based navigators (3D self-iNAVs) at multiple respiratory and/or cardiac phases are reconstructed from binned data. The displacement fields are estimated by comparing these 3D self-iNAVs using an established image registration method. A series of linear operators, each modeling data at a particular state of the heart, are constructed using image-space gridding with the estimated displacement fields. By vertically stacking both data and models from different states, an optimization problem that incorporates non-rigid motion is formulated and subsequently solved with an iterative solver, yielding the non-rigid motion-corrected reconstruction.

ISBN: 9798382231112Subjects--Topical Terms:

535387
Biomedical engineering.
Subjects--Index Terms:

Angiography

Tools for Coronary MR Angiography: Non-Cartesian Trajectories and Motion-Corrected Reconstructions.
LDR:05161nmm a2200373 4500 001 2403661
005 20241118135906.5
006 m o d
007 cr#unu||||||||
008 251215s2024 ||||||||||||||||| ||eng d
020 $a 9798382231112
035 $a (MiAaPQ)AAI31255811
035 $a (MiAaPQ)sc219dv4593
035 $a AAI31255811
040 $a MiAaPQ $c MiAaPQ
100 1 $a Kwang Eun Jang. $3 3773931
245 1 0 $a Tools for Coronary MR Angiography: Non-Cartesian Trajectories and Motion-Corrected Reconstructions.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 129 p.
500 $a Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
500 $a Advisor: Dwight George Nishimura.
502 $a Thesis (Ph.D.)--Stanford University, 2024.
520 $a X-ray coronary angiography remains the predominant technique for coronary artery examination, yet it exposes both patients and operators to ionizing radiation and is invasive. Despite strong demand for non-invasive and safer alternatives, the clinical adoption of coronary MR angiography has been slow. The primary challenges in coronary MR angiography are its prolonged scan time and limited resolution. Because coronary arteries often follow tortuous paths and have relatively small diameters, high resolving power is required. Additionally, whole-heart imaging is preferred over target-volume methods, as it reduces operator dependency. Achieving large volumetric coverage at high resolution demands collecting a significant amount of data. However, continuous myocardial movement restricts data collection to specific cardiac cycles, necessitating the segmented data acquisition over multiple heartbeats. Consequently, coronary MR angiography is slow and susceptible to both respiratory and cardiac motion. This dissertation introduces several tools for coronary MR angiography categorized into non-Cartesian trajectory design and non-rigid motion-corrected reconstruction. Non-Cartesian trajectories often exhibit higher sampling efficiency and greater robustness against undersampling compared to traditional rectilinear approaches, making them particularly useful for reducing the overall scan duration. Because non-rigid models capture a wide range of transformations, non-rigid motion-corrected reconstruction reduces resolution loss due to respiratory and cardiac motion.More specifically, this thesis introduces a unified methodology for designing three commonly used non-Cartesian trajectories: 3D Radial, 3D Cones, and Stack-of-spirals. The fundamental idea of the proposed method is that a non-Cartesian trajectory can be regarded as a discretized version of an analytic coordinate defined with a set of template trajectories. In the proposed method, discretizing the analytic coordinate is equivalent to identifying rotation angles for template trajectories, yielding the non-Cartesian trajectory. The discretization is achieved by creating a spiral path on a surface and sampling points along this path. This approach also allows for the analytical computation of the density compensation factor and the incorporation of useful features such as variable-density and spherical design.Additionally, this dissertation introduces image-space gridding for representing non-rigid motion. Similar to "gridding," a technique commonly employed to resample non-Cartesian k-space data onto a Cartesian grid, image-space gridding consists of an exact forward-ad joint pair of linear operators that resample an image onto a Cartesian grid warped by a displacement field. Image-space gridding allows incorporating nonrigid transformations into the data acquisition model. This formulation accounts for the inhomogeneity of data collected at different states of a target object; in the context of cardiac MR imaging, this formulation accounts for non-rigid transformations of the heart during the segmented data acquisition over multiple heartbeats.As a demonstration, this thesis presents non-rigid motion-corrected reconstructions for free-breathing coronary MR angiography. Following translational motion correction, 3D self-navigating image-based navigators (3D self-iNAVs) at multiple respiratory and/or cardiac phases are reconstructed from binned data. The displacement fields are estimated by comparing these 3D self-iNAVs using an established image registration method. A series of linear operators, each modeling data at a particular state of the heart, are constructed using image-space gridding with the estimated displacement fields. By vertically stacking both data and models from different states, an optimization problem that incorporates non-rigid motion is formulated and subsequently solved with an iterative solver, yielding the non-rigid motion-corrected reconstruction.
590 $a School code: 0212.
650 4 $a Biomedical engineering. $3 535387
650 4 $a Medical imaging. $3 3172799
653 $a Angiography
653 $a Coronary artery
653 $a Whole-heart imaging
653 $a Non-Cartesian trajectories
690 $a 0541
690 $a 0574
710 2 $a Stanford University. $3 754827
773 0 $t Dissertations Abstracts International $g 85-11B.
790 $a 0212
791 $a Ph.D.
792 $a 2024
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31255811