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

MRI-Guided and Compressed Sensing Re...

Mersereau, Bryant Garrison.

FindBook

Google Book

Amazon

博客來

MRI-Guided and Compressed Sensing Reconstruction Methods for PET/MRI.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	MRI-Guided and Compressed Sensing Reconstruction Methods for PET/MRI./
作者:	Mersereau, Bryant Garrison.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2017,
面頁冊數:	118 p.
附註:	Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
Contained By:	Dissertation Abstracts International78-10B(E).
標題:	Medical imaging. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10610770
ISBN:	9781369856538

MRI-Guided and Compressed Sensing Reconstruction Methods for PET/MRI.
Mersereau, Bryant Garrison.

MRI-Guided and Compressed Sensing Reconstruction Methods for PET/MRI. - Ann Arbor : ProQuest Dissertations & Theses, 2017 - 118 p.

Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.

Thesis (Ph.D.)--North Carolina State University, 2017.

Recent advances in hybrid positron emission tomography/magnetic resonance imaging (PET/MRI) have opened up a multitude of research areas in neurological, oncological, and cardiac imaging. Using a PET/MRI scanner, complementary functional information from PET and structural information from MRI can be acquired simultaneously. One technical advantage of this hybrid imaging modality is the opportunity to use MRI information to improve the quality of PET reconstruction.

ISBN: 9781369856538Subjects--Topical Terms:

3172799
Medical imaging.

MRI-Guided and Compressed Sensing Reconstruction Methods for PET/MRI.
LDR:03510nmm a2200325 4500 001 2122746
005 20170922124943.5
008 180830s2017 ||||||||||||||||| ||eng d
020 $a 9781369856538
035 $a (MiAaPQ)AAI10610770
035 $a AAI10610770
040 $a MiAaPQ $c MiAaPQ
100 1 $a Mersereau, Bryant Garrison. $3 3284707
245 1 0 $a MRI-Guided and Compressed Sensing Reconstruction Methods for PET/MRI.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2017
300 $a 118 p.
500 $a Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
500 $a Adviser: David S. Lalush.
502 $a Thesis (Ph.D.)--North Carolina State University, 2017.
520 $a Recent advances in hybrid positron emission tomography/magnetic resonance imaging (PET/MRI) have opened up a multitude of research areas in neurological, oncological, and cardiac imaging. Using a PET/MRI scanner, complementary functional information from PET and structural information from MRI can be acquired simultaneously. One technical advantage of this hybrid imaging modality is the opportunity to use MRI information to improve the quality of PET reconstruction.
520 $a There are two major challenges to MR-guided PET reconstruction. The first is the computational challenge of incorporating MR information into iterative PET reconstruction. Serial CPU-based implementations of the OSEM PET reconstruction algorithm can take up to 30 minutes to complete, which can be stressful on research and clinical workflows. To overcome this obstacle, the OSEM algorithm was parallelized for GPU processing using the CUDA programming language. The parallelized reconstruction program has been shown to reconstruct PET images 60 times faster than a serial implementation of the OSEM algorithm.
520 $a The second challenge is that acquired PET/MRI datasets can be problematic to build due to the high logistical and monetary cost associated with recruiting and scanning patients. To alleviate this problem a digital PET simulation platform was developed. This platform utilizes PET phantoms built from patient MRI and CT images to simulate realistic PET images. The platform has shown to produce results consistent with acquired PET data and provides high configurability and flexibility as a simulation tool.
520 $a While PET provides a unique diagnostic imaging utility, it suffers from a host of physical and reconstruction effects which degrade its image quality. An MRI-guided PET reconstruction method was developed for making PET less susceptible to partial volume effects. The method incorporates an MRI prior comprised of multiple MRI images directly into the PET reconstruction algorithm and reconstructs PET images at MRI scale. Using the prior and MAP reconstruction criterion, noise was successfully reduced in a spatially-varying manner in the resulting PET images, and the anatomical guidance of the prior was demonstrated.
520 $a Compressed sensing is a relatively new concept in MRI imaging being used to accelerate acquisitions. For PET/MRI imaging this could mean better image simultaneity between the modalities or motion compensation for PET. A variety of sparse transforms were investigated as well as the spatiotemporal smoothing capabilities of compressed sensing.
590 $a School code: 0155.
650 4 $a Medical imaging. $3 3172799
690 $a 0574
710 2 $a North Carolina State University. $b Biomedical Engineering. $3 3284708
773 0 $t Dissertation Abstracts International $g 78-10B(E).
790 $a 0155
791 $a Ph.D.
792 $a 2017
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10610770