GPU-Accelerated Bi-Objective Treatment Planning for Prostate High-Dose-Rate Brachytherapy.

PURPOSE The purpose of this study is to improve upon a recently introduced bi-objective treatment planning method for prostate HDR brachytherapy, both in terms of resulting plan quality and runtime requirements, to the extent that its execution time is clinically acceptable. METHODS Bi-objective treatment planning is done using a state-of-the-art multi-objective evolutionary algorithm, which produces a large number of potential treatment plans with different trade-offs between coverage of the target volumes and sparing organs at risk. A Graphics Processing Unit (GPU) is used for large-scale parallelization of dose calculations and the calculation of the Dose-Volume (DV) indices of potential treatment plans. Moreover, the objectives of the previously used bi-objective optimization model are modified to produce better results. RESULTS We applied the GPU-accelerated bi-objective treatment planning method to a set of 18 patients, resulting in a set containing a few hundred potential treatment plans with different trade-offs for each of these patients. Due to accelerations introduced in this article, results previously achieved after 1 hour are now achieved within 30 seconds of optimization. We found plans satisfying the clinical protocol for 15 out of 18 patients, whereas this was the case for only 4 out of 18 clinical plans. Higher quality treatment plans are obtained when the accuracy of DV index calculation is increased using more dose calculation points, requiring still no more than 3 minutes of optimization for 100,000 points. CONCLUSIONS Large sets of high-quality treatment plans that trade off coverage and sparing are now achievable within 30 seconds, due to the GPU-acceleration of a previously introduced bi-objective treatment planning method for prostate HDR brachytherapy. Higher quality plans can be achieved when optimizing for 3 minutes, which we still consider to be clinically acceptable. This allows for more insightful treatment plan selection in a clinical setting. This article is protected by copyright. All rights reserved.

[1]  N. H. Luong,et al.  Evaluation of bi-objective treatment planning for high-dose-rate prostate brachytherapy-A retrospective observer study. , 2019, Brachytherapy.

[2]  R. K. Ursem.  Multi-objective Optimization using Evolutionary Algorithms , 2009 .

[3]  Tian-Li Yu,et al.  Speeding up DSMGA-II on CUDA platform , 2017, GECCO.

[4]  J. Williamson,et al.  Update of AAPM Task Group No. 43 Report: A revised AAPM protocol for brachytherapy dose calculations. , 2003, Medical physics.

[5]  Dimos Baltas,et al.  40 HIPO: A hybrid inverse treatment planning optimization algorithm in HDR brachytherapy , 2005 .

[6]  Ivan Merelli,et al.  A Fine-Grained CUDA Implementation of the Multi-objective Evolutionary Approach NSGA-II: Potential Impact for Computational and Systems Biology Applications , 2014, CIBB.

[7]  Ali S. Meigooni,et al.  Introduction and Innovations in Brachytherapy , 2012 .

[8]  Steve B. Jiang,et al.  Moving GPU‐OpenCL‐based Monte Carlo dose calculation toward clinical use: Automatic beam commissioning and source sampling for treatment plan dose calculation , 2017, Journal of applied clinical medical physics.

[9]  Ken Goldberg,et al.  Optimization of HDR brachytherapy dose distributions using linear programming with penalty costs. , 2006, Medical physics.

[10]  J Pouliot,et al.  Inverse planning anatomy-based dose optimization for HDR-brachytherapy of the prostate using fast simulated annealing algorithm and dedicated objective function. , 2001, Medical physics.

[11]  Leocadio G. Casado,et al.  Non-dominated sorting procedure for Pareto dominance ranking on multicore CPU and/or GPU , 2016, Journal of Global Optimization.

[12]  A Bel,et al.  Accelerated ray tracing for radiotherapy dose calculations on a GPU. , 2009, Medical physics.

[13]  Man Leung Wong,et al.  Parallel multi-objective evolutionary algorithms on graphics processing units , 2009, GECCO '09.

[14]  Philippe Després,et al.  A multi-criteria optimization approach for HDR prostate brachytherapy: I. Pareto surface approximation , 2018, Physics in medicine and biology.

[15]  Peter A. N. Bosman,et al.  Application and benchmarking of multi-objective evolutionary algorithms on high-dose-rate brachytherapy planning for prostate cancer treatment , 2017, Swarm Evol. Comput..

[16]  Christian Kirisits,et al.  Accuracy of volume and DVH parameters determined with different brachytherapy treatment planning systems. , 2007, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[17]  J. Pouliot,et al.  American Brachytherapy Society consensus guidelines for high-dose-rate prostate brachytherapy. , 2012, Brachytherapy.

[18]  Peter A. N. Bosman,et al.  Large-scale parallelization of partial evaluations in evolutionary algorithms for real-world problems , 2018, GECCO.

[19]  V. Patera,et al.  GPU-accelerated Monte Carlo Code for Fast Dose Recalculation in Proton Beam Therapy , 2017 .

[20]  M. Ranney,et al.  Beyond the bedside: Clinicians as guardians of public health, medicine and science , 2020, The American Journal of Emergency Medicine.

[21]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[22]  Peter A. N. Bosman,et al.  PV-0256: Fast and insightful bi-objective HDR prostate brachytherapy planning , 2018 .

[23]  Peter A. N. Bosman,et al.  The multi-objective real-valued gene-pool optimal mixing evolutionary algorithm , 2017, GECCO.

[24]  Nathan Bell,et al.  Thrust: A Productivity-Oriented Library for CUDA , 2012 .

[25]  J. Williamson,et al.  Update of AAPM Task Group No. 43 Report: A revised AAPM protocol for brachytherapy dose calculations. , 2004 .

[26]  Alper Atamtürk,et al.  IPIP: A new approach to inverse planning for HDR brachytherapy by directly optimizing dosimetric indices. , 2010, Medical physics.

[27]  Benoît Ozell,et al.  GPUMCD: A new GPU-oriented Monte Carlo dose calculation platform. , 2011, Medical physics.

[28]  T M Deist,et al.  High-dose-rate prostate brachytherapy inverse planning on dose-volume criteria by simulated annealing. , 2016, Physics in medicine and biology.

[29]  D. den Hertog,et al.  Mixed integer programming improves comprehensibility and plan quality in inverse optimization of prostate HDR brachytherapy , 2013, Physics in medicine and biology.

[30]  A. Niemierko,et al.  Random sampling for evaluating treatment plans. , 1990, Medical physics.

[31]  Lei Xing,et al.  GPU computing in medical physics: a review. , 2011, Medical physics.