A hybrid evolutionary algorithm for multi-objective anatomy-based dose optimization in high-dose-rate brachytherapy

Lahanas, M; Baltas, D; Zamboglou, N

dc.contributor.author	Lahanas, M	en
dc.contributor.author	Baltas, D	en
dc.contributor.author	Zamboglou, N	en
dc.date.accessioned	2014-03-01T01:18:32Z
dc.date.available	2014-03-01T01:18:32Z
dc.date.issued	2003	en
dc.identifier.issn	0031-9155	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/15065
dc.subject.classification	Engineering, Biomedical	en
dc.subject.classification	Radiology, Nuclear Medicine & Medical Imaging	en
dc.subject.other	Evolutionary algorithms	en
dc.subject.other	Gradient methods	en
dc.subject.other	Optimization	en
dc.subject.other	Simulated annealing	en
dc.subject.other	Brachytherapy	en
dc.subject.other	Biomedical engineering	en
dc.subject.other	algorithm	en
dc.subject.other	anatomy	en
dc.subject.other	article	en
dc.subject.other	brachytherapy	en
dc.subject.other	controlled study	en
dc.subject.other	decision making	en
dc.subject.other	histogram	en
dc.subject.other	implant	en
dc.subject.other	priority journal	en
dc.subject.other	process optimization	en
dc.subject.other	prostate	en
dc.subject.other	radiation dose	en
dc.subject.other	radiation dose distribution	en
dc.subject.other	simulation	en
dc.subject.other	statistical analysis	en
dc.subject.other	Algorithms	en
dc.subject.other	Brachytherapy	en
dc.subject.other	Dose-Response Relationship, Radiation	en
dc.subject.other	Humans	en
dc.subject.other	Male	en
dc.subject.other	Models, Biological	en
dc.subject.other	Organ Specificity	en
dc.subject.other	Prostatic Neoplasms	en
dc.subject.other	Quality Control	en
dc.subject.other	Radiation Injuries	en
dc.subject.other	Radiometry	en
dc.subject.other	Radiotherapy Dosage	en
dc.subject.other	Radiotherapy Planning, Computer-Assisted	en
dc.subject.other	Sensitivity and Specificity	en
dc.title	A hybrid evolutionary algorithm for multi-objective anatomy-based dose optimization in high-dose-rate brachytherapy	en
heal.type	journalArticle	en
heal.identifier.primary	10.1088/0031-9155/48/3/309	en
heal.identifier.secondary	http://dx.doi.org/10.1088/0031-9155/48/3/309	en
heal.language	English	en
heal.publicationDate	2003	en
heal.abstract	Multiple objectives must be considered in anatomy-based dose optimization for high-dose-rate brachytherapy and a large number of parameters must be optimized to satisfy often competing objectives. For objectives expressed solely in terms of dose variances, deterministic gradient-based algorithms can be applied and a weighted sum approach is able to produce a representative set of non-dominated solutions. As the number of objectives increases, or non-convex objectives are used, local minima can be present and deterministic or stochastic algorithms such as simulated annealing either cannot be used or are not efficient. In this case we employ a modified hybrid version of the multi-objective optimization algorithm NSGA-II. This, in combination with the deterministic optimization algorithm, produces a representative sample of the Pareto set. This algorithm can be used with any kind of objectives, including non-convex, and does not require artificial importance factors. A representation of the trade-off surface can be obtained with more than 1000 non-dominated solutions in 2-5 min. An analysis of the solutions provides information on the possibilities available using these objectives. Simple decision making tools allow the selection of a solution that provides a best fit for the clinical goals. We show an example with a prostate implant and compare results obtained by variance and dose-volume histogram (DVH) based objectives.	en
heal.publisher	IOP PUBLISHING LTD	en
heal.journalName	Physics in Medicine and Biology	en
dc.identifier.doi	10.1088/0031-9155/48/3/309	en
dc.identifier.isi	ISI:000181367300009	en
dc.identifier.volume	48	en
dc.identifier.issue	3	en
dc.identifier.spage	399	en
dc.identifier.epage	415	en