Loading…

A GPU-based multi-criteria optimization algorithm for HDR brachytherapy

Currently in HDR brachytherapy planning, a manual fine-tuning of an objective function is necessary to obtain case-specific valid plans. This study intends to facilitate this process by proposing a patient-specific inverse planning algorithm for HDR prostate brachytherapy: GPU-based multi-criteria o...

Full description

Saved in:

Bibliographic Details
Published in:	Physics in medicine & biology 2019-05, Vol.64 (10), p.105005-105005
Main Authors:	Bélanger, Cédric, Cui, Songye, Ma, Yunzhi, Després, Philippe, Adam M Cunha, J, Beaulieu, Luc
Format:	Article
Language:	English
Subjects:	Algorithms brachytherapy Brachytherapy - methods Brachytherapy - standards GPU Humans Male optimization patient-specific prostate cancer Prostatic Neoplasms - radiotherapy Radiotherapy Dosage Radiotherapy Planning, Computer-Assisted - methods Radiotherapy Planning, Computer-Assisted - standards treatment planning
Citations:	Items that this one cites Items that cite this one
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

cited_by	cdi_FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93
cites	cdi_FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93
container_end_page	105005
container_issue	10
container_start_page	105005
container_title	Physics in medicine & biology
container_volume	64
creator	Bélanger, Cédric Cui, Songye Ma, Yunzhi Després, Philippe Adam M Cunha, J Beaulieu, Luc
description	Currently in HDR brachytherapy planning, a manual fine-tuning of an objective function is necessary to obtain case-specific valid plans. This study intends to facilitate this process by proposing a patient-specific inverse planning algorithm for HDR prostate brachytherapy: GPU-based multi-criteria optimization (gMCO). Two GPU-based optimization engines including simulated annealing (gSA) and a quasi-Newton optimizer (gL-BFGS) were implemented to compute multiple plans in parallel. After evaluating the equivalence and the computation performance of these two optimization engines, one preferred optimization engine was selected for the gMCO algorithm. Five hundred sixty-two previously treated prostate HDR cases were divided into validation set (100) and test set (462). In the validation set, the number of Pareto optimal plans to achieve the best plan quality was determined for the gMCO algorithm. In the test set, gMCO plans were compared with the physician-approved clinical plans. Our results indicated that the optimization process is equivalent between gL-BFGS and gSA, and that the computational performance of gL-BFGS is up to 67 times faster than gSA. Over 462 cases, the number of clinically valid plans was 428 (92.6%) for clinical plans and 461 (99.8%) for gMCO plans. The number of valid plans with target coverage greater than 95% was 288 (62.3%) for clinical plans and 414 (89.6%) for gMCO plans. The mean planning time was 9.4 s for the gMCO algorithm to generate 1000 Pareto optimal plans. In conclusion, gL-BFGS is able to compute thousands of SA equivalent treatment plans within a short time frame. Powered by gL-BFGS, an ultra-fast and robust multi-criteria optimization algorithm was implemented for HDR prostate brachytherapy. Plan pools with various trade-offs can be created with this algorithm. A large-scale comparison against physician approved clinical plans showed that treatment plan quality could be improved and planning time could be significantly reduced with the proposed gMCO algorithm.
doi_str_mv	10.1088/1361-6560/ab1817
format	article
fullrecord	<record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmed_primary_30970341</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2207937241</sourcerecordid><originalsourceid>FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93</originalsourceid><addsrcrecordid>eNp1kD1PwzAQhi0EoqWwM6GMDITexc7XWBVokSqBEJ0t23Goq6QOdjKUX0-qlG5MJ9097yvdQ8gtwiNClk2RJhgmcQJTITHD9IyMT6tzMgagGOYYxyNy5f0WADGL2CUZUchToAzHZDELFu_rUAqvi6DuqtaEyplWOyMC27SmNj-iNXYXiOrL9odNHZTWBcunj0A6oTb7dqOdaPbX5KIUldc3xzkh65fnz_kyXL0tXuezVahoztowxbgok7jEEqnMo0gryRIGoAtNZZYXKlaFSPpzDMBQIaOoEYWkpcZE6JxOyP3Q2zj73Wnf8tp4patK7LTtPI8iSHOaRgx7FAZUOeu90yVvnKmF23MEftDHD674wRUf9PWRu2N7J2tdnAJ_vnrgYQCMbfjWdm7XP_t_3y_wkXgO</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2207937241</pqid></control><display><type>article</type><title>A GPU-based multi-criteria optimization algorithm for HDR brachytherapy</title><source>Institute of Physics:Jisc Collections:IOP Publishing Read and Publish 2024-2025 (Reading List)</source><creator>Bélanger, Cédric ; Cui, Songye ; Ma, Yunzhi ; Després, Philippe ; Adam M Cunha, J ; Beaulieu, Luc</creator><creatorcontrib>Bélanger, Cédric ; Cui, Songye ; Ma, Yunzhi ; Després, Philippe ; Adam M Cunha, J ; Beaulieu, Luc</creatorcontrib><description>Currently in HDR brachytherapy planning, a manual fine-tuning of an objective function is necessary to obtain case-specific valid plans. This study intends to facilitate this process by proposing a patient-specific inverse planning algorithm for HDR prostate brachytherapy: GPU-based multi-criteria optimization (gMCO). Two GPU-based optimization engines including simulated annealing (gSA) and a quasi-Newton optimizer (gL-BFGS) were implemented to compute multiple plans in parallel. After evaluating the equivalence and the computation performance of these two optimization engines, one preferred optimization engine was selected for the gMCO algorithm. Five hundred sixty-two previously treated prostate HDR cases were divided into validation set (100) and test set (462). In the validation set, the number of Pareto optimal plans to achieve the best plan quality was determined for the gMCO algorithm. In the test set, gMCO plans were compared with the physician-approved clinical plans. Our results indicated that the optimization process is equivalent between gL-BFGS and gSA, and that the computational performance of gL-BFGS is up to 67 times faster than gSA. Over 462 cases, the number of clinically valid plans was 428 (92.6%) for clinical plans and 461 (99.8%) for gMCO plans. The number of valid plans with target coverage greater than 95% was 288 (62.3%) for clinical plans and 414 (89.6%) for gMCO plans. The mean planning time was 9.4 s for the gMCO algorithm to generate 1000 Pareto optimal plans. In conclusion, gL-BFGS is able to compute thousands of SA equivalent treatment plans within a short time frame. Powered by gL-BFGS, an ultra-fast and robust multi-criteria optimization algorithm was implemented for HDR prostate brachytherapy. Plan pools with various trade-offs can be created with this algorithm. A large-scale comparison against physician approved clinical plans showed that treatment plan quality could be improved and planning time could be significantly reduced with the proposed gMCO algorithm.</description><identifier>ISSN: 0031-9155</identifier><identifier>ISSN: 1361-6560</identifier><identifier>EISSN: 1361-6560</identifier><identifier>DOI: 10.1088/1361-6560/ab1817</identifier><identifier>PMID: 30970341</identifier><identifier>CODEN: PHMBA7</identifier><language>eng</language><publisher>England: IOP Publishing</publisher><subject>Algorithms ; brachytherapy ; Brachytherapy - methods ; Brachytherapy - standards ; GPU ; Humans ; Male ; optimization ; patient-specific ; prostate cancer ; Prostatic Neoplasms - radiotherapy ; Radiotherapy Dosage ; Radiotherapy Planning, Computer-Assisted - methods ; Radiotherapy Planning, Computer-Assisted - standards ; treatment planning</subject><ispartof>Physics in medicine & biology, 2019-05, Vol.64 (10), p.105005-105005</ispartof><rights>2019 Institute of Physics and Engineering in Medicine</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93</citedby><cites>FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93</cites><orcidid>0000-0003-0429-6366 ; 0000-0002-4163-7353 ; 0000-0003-2455-0376 ; 0000-0002-0142-9277</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>315,786,790,27957,27958</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/30970341$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Bélanger, Cédric</creatorcontrib><creatorcontrib>Cui, Songye</creatorcontrib><creatorcontrib>Ma, Yunzhi</creatorcontrib><creatorcontrib>Després, Philippe</creatorcontrib><creatorcontrib>Adam M Cunha, J</creatorcontrib><creatorcontrib>Beaulieu, Luc</creatorcontrib><title>A GPU-based multi-criteria optimization algorithm for HDR brachytherapy</title><title>Physics in medicine & biology</title><addtitle>PMB</addtitle><addtitle>Phys. Med. Biol</addtitle><description>Currently in HDR brachytherapy planning, a manual fine-tuning of an objective function is necessary to obtain case-specific valid plans. This study intends to facilitate this process by proposing a patient-specific inverse planning algorithm for HDR prostate brachytherapy: GPU-based multi-criteria optimization (gMCO). Two GPU-based optimization engines including simulated annealing (gSA) and a quasi-Newton optimizer (gL-BFGS) were implemented to compute multiple plans in parallel. After evaluating the equivalence and the computation performance of these two optimization engines, one preferred optimization engine was selected for the gMCO algorithm. Five hundred sixty-two previously treated prostate HDR cases were divided into validation set (100) and test set (462). In the validation set, the number of Pareto optimal plans to achieve the best plan quality was determined for the gMCO algorithm. In the test set, gMCO plans were compared with the physician-approved clinical plans. Our results indicated that the optimization process is equivalent between gL-BFGS and gSA, and that the computational performance of gL-BFGS is up to 67 times faster than gSA. Over 462 cases, the number of clinically valid plans was 428 (92.6%) for clinical plans and 461 (99.8%) for gMCO plans. The number of valid plans with target coverage greater than 95% was 288 (62.3%) for clinical plans and 414 (89.6%) for gMCO plans. The mean planning time was 9.4 s for the gMCO algorithm to generate 1000 Pareto optimal plans. In conclusion, gL-BFGS is able to compute thousands of SA equivalent treatment plans within a short time frame. Powered by gL-BFGS, an ultra-fast and robust multi-criteria optimization algorithm was implemented for HDR prostate brachytherapy. Plan pools with various trade-offs can be created with this algorithm. A large-scale comparison against physician approved clinical plans showed that treatment plan quality could be improved and planning time could be significantly reduced with the proposed gMCO algorithm.</description><subject>Algorithms</subject><subject>brachytherapy</subject><subject>Brachytherapy - methods</subject><subject>Brachytherapy - standards</subject><subject>GPU</subject><subject>Humans</subject><subject>Male</subject><subject>optimization</subject><subject>patient-specific</subject><subject>prostate cancer</subject><subject>Prostatic Neoplasms - radiotherapy</subject><subject>Radiotherapy Dosage</subject><subject>Radiotherapy Planning, Computer-Assisted - methods</subject><subject>Radiotherapy Planning, Computer-Assisted - standards</subject><subject>treatment planning</subject><issn>0031-9155</issn><issn>1361-6560</issn><issn>1361-6560</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNp1kD1PwzAQhi0EoqWwM6GMDITexc7XWBVokSqBEJ0t23Goq6QOdjKUX0-qlG5MJ9097yvdQ8gtwiNClk2RJhgmcQJTITHD9IyMT6tzMgagGOYYxyNy5f0WADGL2CUZUchToAzHZDELFu_rUAqvi6DuqtaEyplWOyMC27SmNj-iNXYXiOrL9odNHZTWBcunj0A6oTb7dqOdaPbX5KIUldc3xzkh65fnz_kyXL0tXuezVahoztowxbgok7jEEqnMo0gryRIGoAtNZZYXKlaFSPpzDMBQIaOoEYWkpcZE6JxOyP3Q2zj73Wnf8tp4patK7LTtPI8iSHOaRgx7FAZUOeu90yVvnKmF23MEftDHD674wRUf9PWRu2N7J2tdnAJ_vnrgYQCMbfjWdm7XP_t_3y_wkXgO</recordid><startdate>20190508</startdate><enddate>20190508</enddate><creator>Bélanger, Cédric</creator><creator>Cui, Songye</creator><creator>Ma, Yunzhi</creator><creator>Després, Philippe</creator><creator>Adam M Cunha, J</creator><creator>Beaulieu, Luc</creator><general>IOP Publishing</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0003-0429-6366</orcidid><orcidid>https://orcid.org/0000-0002-4163-7353</orcidid><orcidid>https://orcid.org/0000-0003-2455-0376</orcidid><orcidid>https://orcid.org/0000-0002-0142-9277</orcidid></search><sort><creationdate>20190508</creationdate><title>A GPU-based multi-criteria optimization algorithm for HDR brachytherapy</title><author>Bélanger, Cédric ; Cui, Songye ; Ma, Yunzhi ; Després, Philippe ; Adam M Cunha, J ; Beaulieu, Luc</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Algorithms</topic><topic>brachytherapy</topic><topic>Brachytherapy - methods</topic><topic>Brachytherapy - standards</topic><topic>GPU</topic><topic>Humans</topic><topic>Male</topic><topic>optimization</topic><topic>patient-specific</topic><topic>prostate cancer</topic><topic>Prostatic Neoplasms - radiotherapy</topic><topic>Radiotherapy Dosage</topic><topic>Radiotherapy Planning, Computer-Assisted - methods</topic><topic>Radiotherapy Planning, Computer-Assisted - standards</topic><topic>treatment planning</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bélanger, Cédric</creatorcontrib><creatorcontrib>Cui, Songye</creatorcontrib><creatorcontrib>Ma, Yunzhi</creatorcontrib><creatorcontrib>Després, Philippe</creatorcontrib><creatorcontrib>Adam M Cunha, J</creatorcontrib><creatorcontrib>Beaulieu, Luc</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Physics in medicine & biology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Bélanger, Cédric</au><au>Cui, Songye</au><au>Ma, Yunzhi</au><au>Després, Philippe</au><au>Adam M Cunha, J</au><au>Beaulieu, Luc</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A GPU-based multi-criteria optimization algorithm for HDR brachytherapy</atitle><jtitle>Physics in medicine & biology</jtitle><stitle>PMB</stitle><addtitle>Phys. Med. Biol</addtitle><date>2019-05-08</date><risdate>2019</risdate><volume>64</volume><issue>10</issue><spage>105005</spage><epage>105005</epage><pages>105005-105005</pages><issn>0031-9155</issn><issn>1361-6560</issn><eissn>1361-6560</eissn><coden>PHMBA7</coden><notes>Institute of Physics and Engineering in Medicine</notes><notes>PMB-108333.R1</notes><notes>ObjectType-Article-1</notes><notes>SourceType-Scholarly Journals-1</notes><notes>ObjectType-Feature-2</notes><notes>content type line 23</notes><abstract>Currently in HDR brachytherapy planning, a manual fine-tuning of an objective function is necessary to obtain case-specific valid plans. This study intends to facilitate this process by proposing a patient-specific inverse planning algorithm for HDR prostate brachytherapy: GPU-based multi-criteria optimization (gMCO). Two GPU-based optimization engines including simulated annealing (gSA) and a quasi-Newton optimizer (gL-BFGS) were implemented to compute multiple plans in parallel. After evaluating the equivalence and the computation performance of these two optimization engines, one preferred optimization engine was selected for the gMCO algorithm. Five hundred sixty-two previously treated prostate HDR cases were divided into validation set (100) and test set (462). In the validation set, the number of Pareto optimal plans to achieve the best plan quality was determined for the gMCO algorithm. In the test set, gMCO plans were compared with the physician-approved clinical plans. Our results indicated that the optimization process is equivalent between gL-BFGS and gSA, and that the computational performance of gL-BFGS is up to 67 times faster than gSA. Over 462 cases, the number of clinically valid plans was 428 (92.6%) for clinical plans and 461 (99.8%) for gMCO plans. The number of valid plans with target coverage greater than 95% was 288 (62.3%) for clinical plans and 414 (89.6%) for gMCO plans. The mean planning time was 9.4 s for the gMCO algorithm to generate 1000 Pareto optimal plans. In conclusion, gL-BFGS is able to compute thousands of SA equivalent treatment plans within a short time frame. Powered by gL-BFGS, an ultra-fast and robust multi-criteria optimization algorithm was implemented for HDR prostate brachytherapy. Plan pools with various trade-offs can be created with this algorithm. A large-scale comparison against physician approved clinical plans showed that treatment plan quality could be improved and planning time could be significantly reduced with the proposed gMCO algorithm.</abstract><cop>England</cop><pub>IOP Publishing</pub><pmid>30970341</pmid><doi>10.1088/1361-6560/ab1817</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0003-0429-6366</orcidid><orcidid>https://orcid.org/0000-0002-4163-7353</orcidid><orcidid>https://orcid.org/0000-0003-2455-0376</orcidid><orcidid>https://orcid.org/0000-0002-0142-9277</orcidid></addata></record>
fulltext	fulltext
identifier	ISSN: 0031-9155
ispartof	Physics in medicine & biology, 2019-05, Vol.64 (10), p.105005-105005
issn	0031-9155 1361-6560 1361-6560
language	eng
recordid	cdi_pubmed_primary_30970341
source	Institute of Physics:Jisc Collections:IOP Publishing Read and Publish 2024-2025 (Reading List)
subjects	Algorithms brachytherapy Brachytherapy - methods Brachytherapy - standards GPU Humans Male optimization patient-specific prostate cancer Prostatic Neoplasms - radiotherapy Radiotherapy Dosage Radiotherapy Planning, Computer-Assisted - methods Radiotherapy Planning, Computer-Assisted - standards treatment planning
title	A GPU-based multi-criteria optimization algorithm for HDR brachytherapy
url	http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-09-22T22%3A29%3A28IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20GPU-based%20multi-criteria%20optimization%20algorithm%20for%20HDR%20brachytherapy&rft.jtitle=Physics%20in%20medicine%20&%20biology&rft.au=B%C3%A9langer,%20C%C3%A9dric&rft.date=2019-05-08&rft.volume=64&rft.issue=10&rft.spage=105005&rft.epage=105005&rft.pages=105005-105005&rft.issn=0031-9155&rft.eissn=1361-6560&rft.coden=PHMBA7&rft_id=info:doi/10.1088/1361-6560/ab1817&rft_dat=%3Cproquest_pubme%3E2207937241%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c394t-715df65f1f13b922ecb46400ede3b89dc5cda6f1f50041c1431e11ab3fe16ae93%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2207937241&rft_id=info:pmid/30970341&rfr_iscdi=true