東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Development of parallel magnetic res...

University of Virginia.

FindBook

Google Book

Amazon

博客來

Development of parallel magnetic resonance imaging methods and their applications in spiral cardiac imaging.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Development of parallel magnetic resonance imaging methods and their applications in spiral cardiac imaging./
作者:	Hu, Peng.
面頁冊數:	108 p.
附註:	Adviser: Craig Meyer.
Contained By:	Dissertation Abstracts International68-11B.
標題:	Biophysics, Medical. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3288359
ISBN:	9780549314240

Development of parallel magnetic resonance imaging methods and their applications in spiral cardiac imaging.
Hu, Peng.

Development of parallel magnetic resonance imaging methods and their applications in spiral cardiac imaging. - 108 p.

Adviser: Craig Meyer.

Thesis (Ph.D.)--University of Virginia, 2008.

In this dissertation, we seek to improve MRI speed by developing a novel method that combines two common fast imaging methods: spiral scanning and parallel imaging. This new method, Parallel Imaging Based On Successive Convolution Operations (BOSCO), is an auto-calibrating parallel imaging method that works well for both Cartesian and non-Cartesian trajectories. BOSCO reconstructs un-aliased images from a linear combination of convolutions of the channel k-space data with small convolution kernels (BOSCO kernels) once the non-Cartesian data is interpolated onto regular grid points using gridding.

ISBN: 9780549314240Subjects--Topical Terms:

1017681
Biophysics, Medical.

Development of parallel magnetic resonance imaging methods and their applications in spiral cardiac imaging.
LDR:03297nam 2200301 a 45 001 862373
005 20100720
008 100720s2008 ||||||||||||||||| ||eng d
020 $a 9780549314240
035 $a (UMI)AAI3288359
035 $a AAI3288359
040 $a UMI $c UMI
100 1 $a Hu, Peng. $3 1030184
245 1 0 $a Development of parallel magnetic resonance imaging methods and their applications in spiral cardiac imaging.
300 $a 108 p.
500 $a Adviser: Craig Meyer.
500 $a Source: Dissertation Abstracts International, Volume: 68-11, Section: B, page: 7185.
502 $a Thesis (Ph.D.)--University of Virginia, 2008.
520 $a In this dissertation, we seek to improve MRI speed by developing a novel method that combines two common fast imaging methods: spiral scanning and parallel imaging. This new method, Parallel Imaging Based On Successive Convolution Operations (BOSCO), is an auto-calibrating parallel imaging method that works well for both Cartesian and non-Cartesian trajectories. BOSCO reconstructs un-aliased images from a linear combination of convolutions of the channel k-space data with small convolution kernels (BOSCO kernels) once the non-Cartesian data is interpolated onto regular grid points using gridding.
520 $a Fundamental theory behind BOSCO and practical issues related to algorithm design and implementation will be covered in detail. Numerous simulation, phantom and in vivo experiments will be conducted to evaluate this technique. The impact of several key parameters of this method on the reconstruction quality will be investigated. After demonstrating the technical developments involved in this technique, we share our initial experience of applying this technique to real time spiral imaging and potentially for real time spiral cardiac stress testing.
520 $a Several features distinguish BOSCO from existing k-space based parallel imaging methods. One is that BOSCO does a gridding first on non-Cartesian data before the actual parallel reconstruction steps. The second is that it uses a small convolution kernel throughout k-space without segmentations that are often used in existing methods. These differences provide several advantages. For example, BOSCO convolution can be performed equivalently in image space as pixel-wise multiplications, which significantly shortens the image reconstruction time. This is very useful for real time imaging, where image reconstruction needs to be fast enough to keep up with the data acquisition at 30 frames per second. For another example, non-Cartesian BOSCO eliminates the need for multiple iterations that is used in non-Cartesian SENSE and the need for performing multiple trainings due to segmentation in k-space.
520 $a Another component of this dissertation is to apply the PILS parallel imaging method to coronary artery imaging. With automatic calculation of coil parameters, i.e. coil center and sensitivity dimension, we show spiral PILS can provide 2X acceleration on coronary scans while maintaining sufficient SNR.
590 $a School code: 0246.
650 4 $a Biophysics, Medical. $3 1017681
690 $a 0760
710 2 $a University of Virginia. $3 645578
773 0 $t Dissertation Abstracts International $g 68-11B.
790 $a 0246
790 1 0 $a Meyer, Craig, $e advisor
791 $a Ph.D.
792 $a 2008
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3288359