東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Automatic image segmentation and tre...

Zhao, Xuan.

Linked to FindBook

Google Book

Amazon

博客來

Automatic image segmentation and treatment planning for radiotherapy.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Automatic image segmentation and treatment planning for radiotherapy./
Author:	Zhao, Xuan.
Description:	149 p.
Notes:	Source: Dissertation Abstracts International, Volume: 75-11(E), Section: B.
Contained By:	Dissertation Abstracts International75-11B(E).
Subject:	Electrical engineering. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3629559
ISBN:	9781321065855

Automatic image segmentation and treatment planning for radiotherapy.
Zhao, Xuan.

Automatic image segmentation and treatment planning for radiotherapy. - 149 p.

Source: Dissertation Abstracts International, Volume: 75-11(E), Section: B.

Thesis (Ph.D.)--Polytechnic Institute of New York University, 2014.

This item must not be sold to any third party vendors.

In this thesis, the problems associated with the automatic, efficient and robust methodologies for breast radiotherapy are considered. Towards this goal, multiple problems are investigated, which includes automatic image segmentation methods for breast and heart, automatic treatment fields planning based on patient-specific geometry and patient-specific optimal treatment setup. Solutions for these problems were developed with an emphasis on accuracy and robustness.

ISBN: 9781321065855Subjects--Topical Terms:

649834
Electrical engineering.

Automatic image segmentation and treatment planning for radiotherapy.
LDR:04238nmm a2200337 4500 001 2059956
005 20150822105424.5
008 170521s2014 ||||||||||||||||| ||eng d
020 $a 9781321065855
035 $a (MiAaPQ)AAI3629559
035 $a AAI3629559
040 $a MiAaPQ $c MiAaPQ
100 1 $a Zhao, Xuan. $3 1949838
245 1 0 $a Automatic image segmentation and treatment planning for radiotherapy.
300 $a 149 p.
500 $a Source: Dissertation Abstracts International, Volume: 75-11(E), Section: B.
500 $a Adviser: Yao Wang.
502 $a Thesis (Ph.D.)--Polytechnic Institute of New York University, 2014.
506 $a This item must not be sold to any third party vendors.
520 $a In this thesis, the problems associated with the automatic, efficient and robust methodologies for breast radiotherapy are considered. Towards this goal, multiple problems are investigated, which includes automatic image segmentation methods for breast and heart, automatic treatment fields planning based on patient-specific geometry and patient-specific optimal treatment setup. Solutions for these problems were developed with an emphasis on accuracy and robustness.
520 $a The first problem is to study and design automatic and robust image segmentation methods for breast and heart in 3-D computer tomography (CT) images. The limitations of conventional segmentation methods are first discussed. Methods of fully automatic localization and segmentation of breast and heart, respectively, are proposed. More specifically, a learning-based method for the whole breast segmentation is proposed to use learned local image statistics from training images to guide the segmentation. Further, a framework for fully automatic heart localization and segmentation is proposed. Statistic models for shape and appearance along the heart boundary are utilized to handle the blurry boundary between heart and liver. We investigate robust parameter estimation in active shape model (ASM) framework and demonstrate its advantage in our application. The incorporation of shape prior into active contour method is also investigated. The merits of the proposed shape-constrained active contour over ASM based methods are demonstrated. Another aspect is robust initialization approaches, as a good initialization is often essential for the accurate segmentation. We investigated in using the geometric relations among organs and global-to-local registration schemes for robustly initialization.
520 $a Another objective of this thesis is to predict the preferred treatment position for specific patient. Previous study has shown that conventional breast radiotherapy in supine position might involve significant amount of surrounding organs in the treatment fields and cause complications to heart and lung. Recently, treatment in prone position has demonstrated reduced dose for certain patients. However, general consensus on the optimal treatment strategy has not been reached. In this study, we investigate how to apply geometric feature-based classifiers to predict the optimal treatment position for individual patient.
520 $a Furthermore, automated and objective method for determining the optimal tangential treatment fields is investigated. We formulate the problem as an optimal linear separation problem and solve it by nding the optimal 3-D plane that separates the whole breast and the included clinical target volumes (CTVs) from the organs at risk (OARs) using a weighted linear Support Vector Machine (SVM). One of the problems is how to incorporate the relative significance of including/avoiding certain organ in the treatment fields. This is achieved by assigning dierent weights for dierent organs based on their significance in the cost function of the SVM. The tangential fields parameters, i.e., the gantry angle, collimator angle, and posterior jaw size of the tangential fields, are derived from the optimal separating plane.
590 $a School code: 1540.
650 4 $a Electrical engineering. $3 649834
650 4 $a Medical imaging. $3 3172799
650 4 $a Engineering. $3 586835
690 $a 0544
690 $a 0574
690 $a 0537
710 2 $a Polytechnic Institute of New York University. $b Electrical and Computer Engineering. $3 1669328
773 0 $t Dissertation Abstracts International $g 75-11B(E).
790 $a 1540
791 $a Ph.D.
792 $a 2014
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3629559