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

Optimization of IMRT using multi-obj...

Tom, Brian C.

FindBook

Google Book

Amazon

博客來

Optimization of IMRT using multi-objective evolutionary algorithms with regularization: A study of complexity vs. deliverability.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Optimization of IMRT using multi-objective evolutionary algorithms with regularization: A study of complexity vs. deliverability./
作者:	Tom, Brian C.
面頁冊數:	214 p.
附註:	Source: Dissertation Abstracts International, Volume: 67-02, Section: B, page: 0962.
Contained By:	Dissertation Abstracts International67-02B.
標題:	Health Sciences, Radiology. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3208276
ISBN:	9780542558696

Optimization of IMRT using multi-objective evolutionary algorithms with regularization: A study of complexity vs. deliverability.
Tom, Brian C.

Optimization of IMRT using multi-objective evolutionary algorithms with regularization: A study of complexity vs. deliverability. - 214 p.

Source: Dissertation Abstracts International, Volume: 67-02, Section: B, page: 0962.

Thesis (Ph.D.)--Rosalind Franklin University of Medicine and Science, 2006.

Intensity Modulated Radiation Therapy (IMRT) has enjoyed success in the clinic by achieving dose escalation to the target while sparing nearby critical structures.

ISBN: 9780542558696Subjects--Topical Terms:

1019076
Health Sciences, Radiology.

Optimization of IMRT using multi-objective evolutionary algorithms with regularization: A study of complexity vs. deliverability.
LDR:03175nmm 2200349 4500 001 1832003
005 20070628102627.5
008 130610s2006 eng d
020 $a 9780542558696
035 $a (UnM)AAI3208276
035 $a AAI3208276
040 $a UnM $c UnM
100 1 $a Tom, Brian C. $3 1920760
245 1 0 $a Optimization of IMRT using multi-objective evolutionary algorithms with regularization: A study of complexity vs. deliverability.
300 $a 214 p.
500 $a Source: Dissertation Abstracts International, Volume: 67-02, Section: B, page: 0962.
500 $a Adviser: John LeVan.
502 $a Thesis (Ph.D.)--Rosalind Franklin University of Medicine and Science, 2006.
520 $a Intensity Modulated Radiation Therapy (IMRT) has enjoyed success in the clinic by achieving dose escalation to the target while sparing nearby critical structures.
520 $a For DMLC plans, regularization is introduced in order to smooth the fluence maps. In this dissertation, regularization is used to smooth the fluence profiles.
520 $a Since SMLC plans have a limited number of intensity levels, smoothing is not a problem. However, in many treatment planning systems, the plans are optimized with beam weights that are continuous. Only after the optimization is complete is when the fluence maps are quantized. This dissertation will study the effects, if any, of quantizing the beam weights.
520 $a In order to study both smoothing DMLC plans and the quantization of SMLC plans, a multi-objective evolutionary algorithm is employed as the optimization method. The main advantages of using these stochastic algorithms is that the beam weights can be represented either in binary or real strings. Clearly, a binary representation is suited for SMLC delivery (discrete intensity levels), while a real representation is more suited for DMLC. Further, in the case of real beam weights, multi-objective evolutionary algorithms can handle conflicting objective functions very well. In fact, regularization can be thought of as having two competing functions: to maintain fidelity to the data, and smoothing the data. The main disadvantage of regularization is the need to specify the regularization parameter, which controls how important the two objectives are relative to one another. Multi-objective evolutionary algorithms do not need such a parameter. In addition, such algorithms yield a set of solutions, each solution representing differing importance factors of the two (or more) objective functions.
520 $a Multi-objective evolutionary algorithms can thus be used to study the effects of quantizing the beam weights for SMLC delivery systems as well studying how regularization can reduce the difference between the optimized and actually delivered plans. This dissertation addresses these two issues. (Abstract shortened by UMI.)
590 $a School code: 1489.
650 4 $a Health Sciences, Radiology. $3 1019076
650 4 $a Physics, Radiation. $3 1019212
650 4 $a Biophysics, Medical. $3 1017681
650 4 $a Health Sciences, Oncology. $3 1018566
690 $a 0574
690 $a 0756
690 $a 0760
690 $a 0992
710 2 0 $a Rosalind Franklin University of Medicine and Science. $3 1025147
773 0 $t Dissertation Abstracts International $g 67-02B.
790 1 0 $a LeVan, John, $e advisor
790 $a 1489
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3208276