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

Real-time four-dimensional cardiac i...

Columbia University.

FindBook

Google Book

Amazon

博客來

Real-time four-dimensional cardiac image analysis for left ventricular function.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Real-time four-dimensional cardiac image analysis for left ventricular function./
作者:	Duan Qi.
面頁冊數:	267 p.
附註:	Adviser: Andrew F. Laine.
Contained By:	Dissertation Abstracts International68-11B.
標題:	Engineering, Biomedical. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoeng/servlet/advanced?query=3290510
ISBN:	9780549339014

Real-time four-dimensional cardiac image analysis for left ventricular function.
Duan Qi.

Real-time four-dimensional cardiac image analysis for left ventricular function. - 267 p.

Adviser: Andrew F. Laine.

Thesis (Ph.D.)--Columbia University, 2008.

In this dissertation, several extensive examples of methods are presented related to image processing, including filter pre-processing, feature extraction, and dynamic analysis.

ISBN: 9780549339014Subjects--Topical Terms:

1017684
Engineering, Biomedical.

Real-time four-dimensional cardiac image analysis for left ventricular function.
LDR:03679nam 2200325 a 45 001 856598
005 20100709
008 100709s2008 ||||||||||||||||| ||eng d
020 $a 9780549339014
035 $a (UMI)AAI3290510
035 $a AAI3290510
040 $a UMI $c UMI
100 1 $a Duan Qi. $3 1023425
245 1 0 $a Real-time four-dimensional cardiac image analysis for left ventricular function.
300 $a 267 p.
500 $a Adviser: Andrew F. Laine.
500 $a Source: Dissertation Abstracts International, Volume: 68-11, Section: B, page: 7470.
502 $a Thesis (Ph.D.)--Columbia University, 2008.
520 $a In this dissertation, several extensive examples of methods are presented related to image processing, including filter pre-processing, feature extraction, and dynamic analysis.
520 $a Cardiovascular disease remains a leading cause of death in the United States. Demand for effective, rapid, and accurate diagnostic tools are of increasing importance. Cardiac imaging analysis can provide tools to visualize the anatomy and function of the heart for diagnosis and treatment of heart disease. Recent advances in 3D and 4D imaging techniques such as real-time three-dimensional echocardiography offer a non-invasive means to capture the complex 4D dynamics of the human heart. However, the tremendous amount of data from these modalities raises challenges to image analyses in term of accuracy, efficiency, and novelty.
520 $a Adaptive filtering by anisotropic diffusion was optimized in an efficient way for real-time performance. This technique can filter a typical real-time 3D ultrasound volume within seconds. As an adaptive filtering technique, important anatomical features such as endo- and epi-cardial borders (edges) and textures can be preserved while allowing noise suppression at the same time. This method served as the preprocessing step for most of RT3DE applications presented in this thesis. In a related pilot study on automated wavelet denoising of parallel cardiac NM data, we showed how acquisition rate can be pushed further by using a multi-scale based denoising technique.
520 $a A novel segmentation framework, using surface function actives (SFA), is proposed in this thesis. As the opposite of level set methods, this framework represents the interface by a representation function with lower dimensionality than the image data. In this way, a tremendous improvement in efficiency of the algorithm can be achieved. Also by utilizing a variational framework, SFA can have similar versatility as a level set framework in terms of expansion with specific energy terms. An example of such an expansion to multi-phase multi-channel level sets was shown through a segmentation software for cardiac MRI.
520 $a To track the motion of the heart, an optical-flow based tracking algorithm was developed and validated on experimental and clinical data for its ability to track ventricular borders. This method was extended for myocardium tracking to estimate the myocardial motion field, from which several metrics of cardiac dynamics could be derived, that someday might serve as a diagnostic tool in clinical practice. Preliminary validation studies showed great correlation in terms of predicted abnormal regions with a physiological finite element model, and high correlation of agreement for strain metrics estimated with cardiac MR tagging.
590 $a School code: 0054.
650 4 $a Engineering, Biomedical. $3 1017684
650 4 $a Health Sciences, Radiology. $3 1019076
690 $a 0541
690 $a 0574
710 2 $a Columbia University. $3 571054
773 0 $t Dissertation Abstracts International $g 68-11B.
790 $a 0054
790 1 0 $a Laine, Andrew F., $e advisor
791 $a Ph.D.
792 $a 2008
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoeng/servlet/advanced?query=3290510