Improving VMAT dose calculation accuracy and planning quality via a GPU-accelerated Fourier transform dose calculation algorithm

IF 2 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Journal of Applied Clinical Medical Physics Pub Date : 2025-02-07 DOI:10.1002/acm2.70002

Kenny Guida, Chaoqiong Ma, Joy Patel, Krishna Reddy, H. Harold Li

{"title":"Improving VMAT dose calculation accuracy and planning quality via a GPU-accelerated Fourier transform dose calculation algorithm","authors":"Kenny Guida, Chaoqiong Ma, Joy Patel, Krishna Reddy, H. Harold Li","doi":"10.1002/acm2.70002","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Background</h3>\n \n <p>Inverse planning typically utilizes fast, less accurate dose calculation algorithms <i>during</i> the iterative optimization process, thus leading to dose calculation errors (DCEs) and suboptimal plans that often require dose normalization and/or plan re-optimization.</p>\n </section>\n \n <section>\n \n <h3> Purpose</h3>\n \n <p>A graphic processing unit (GPU) accelerated Fourier transform dose calculation (FTDC) was recently commissioned at our institution during the Eclipse treatment planning system (Varian Medical Systems) v18.0 upgrade. We hypothesize that FTDC could reduce DCEs and planning failure rates (PFRs) compared to its predecessor, multi-resolution dose calculation (MRDC), while improving efficiency through utilization of GPUs.</p>\n </section>\n \n <section>\n \n <h3> Methods</h3>\n \n <p>Fifty lung SBRT plans were optimized with MRDC and FTDC dose calculation algorithms. Acuros XB (AXB) was then used for final dose calculations. DCEs for target and organ-at-risk (OAR) were calculated as the percent difference between AXB and dose calculated at the final optimization step. Plan quality was assessed using an in-house planning scorecard where PFRs were calculated as the percentage of plans that had a plan score less than 90% with optimal plans scored at 100%.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>FTDC showed excellent agreement with AXB in terms of planning target volume (PTV) coverage, as PTV D95% DCE<sub>FTDC</sub> averaged 0.8% ± 0.9%, compared to DCE<sub>MRDC</sub>’s −2.5% ± 3.2%. DCEs for thoracic OARs were reduced with less variation when optimizing with FTDC as compared to MRDC. FTDC had a PFR of 10% (5 out of 50) versus MRDC's 32% (16 out of 50). The subsequent re-optimization rate resulted from a plan normalization of 3% or greater was 4% for FTDC compared to MRDC's 38%. FTDC with GPU acceleration reduced optimization time by 75% on average compared to MRDC without GPU acceleration.</p>\n </section>\n \n <section>\n \n <h3> Conclusions</h3>\n \n <p>FTDC shows more accurate dose calculation accuracy compared to MRDC. Its use during the optimization process improved planning quality and efficiency assisted with GPUs.</p>\n </section>\n </div>","PeriodicalId":14989,"journal":{"name":"Journal of Applied Clinical Medical Physics","volume":"26 4","pages":""},"PeriodicalIF":2.0000,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/acm2.70002","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Applied Clinical Medical Physics","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/acm2.70002","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Inverse planning typically utilizes fast, less accurate dose calculation algorithms during the iterative optimization process, thus leading to dose calculation errors (DCEs) and suboptimal plans that often require dose normalization and/or plan re-optimization.

Purpose

A graphic processing unit (GPU) accelerated Fourier transform dose calculation (FTDC) was recently commissioned at our institution during the Eclipse treatment planning system (Varian Medical Systems) v18.0 upgrade. We hypothesize that FTDC could reduce DCEs and planning failure rates (PFRs) compared to its predecessor, multi-resolution dose calculation (MRDC), while improving efficiency through utilization of GPUs.

Methods

Fifty lung SBRT plans were optimized with MRDC and FTDC dose calculation algorithms. Acuros XB (AXB) was then used for final dose calculations. DCEs for target and organ-at-risk (OAR) were calculated as the percent difference between AXB and dose calculated at the final optimization step. Plan quality was assessed using an in-house planning scorecard where PFRs were calculated as the percentage of plans that had a plan score less than 90% with optimal plans scored at 100%.

Results

FTDC showed excellent agreement with AXB in terms of planning target volume (PTV) coverage, as PTV D95% DCE_FTDC averaged 0.8% ± 0.9%, compared to DCE_MRDC’s −2.5% ± 3.2%. DCEs for thoracic OARs were reduced with less variation when optimizing with FTDC as compared to MRDC. FTDC had a PFR of 10% (5 out of 50) versus MRDC's 32% (16 out of 50). The subsequent re-optimization rate resulted from a plan normalization of 3% or greater was 4% for FTDC compared to MRDC's 38%. FTDC with GPU acceleration reduced optimization time by 75% on average compared to MRDC without GPU acceleration.

Conclusions

FTDC shows more accurate dose calculation accuracy compared to MRDC. Its use during the optimization process improved planning quality and efficiency assisted with GPUs.

Abstract Image

查看原文本刊更多论文

通过gpu加速傅里叶变换剂量计算算法提高VMAT剂量计算精度和规划质量。

背景：逆规划通常在迭代优化过程中使用快速、不太精确的剂量计算算法，从而导致剂量计算误差（dce）和次优计划，通常需要剂量归一化和/或计划重新优化。目的：在Eclipse治疗计划系统（Varian Medical Systems） v18.0升级期间，我们机构最近委托了一个图形处理单元（GPU）加速傅里叶变换剂量计算（FTDC）。我们假设与之前的多分辨率剂量计算（MRDC）相比，FTDC可以降低dce和规划故障率（PFRs），同时通过利用gpu提高效率。方法：采用MRDC和FTDC剂量计算算法对50个肺SBRT方案进行优化。然后使用acros XB （AXB）进行最终剂量计算。靶器官和危险器官（OAR）的dce计算为最终优化步骤计算的AXB与剂量之间的差异百分比。计划质量是使用内部计划记分卡进行评估的，其中pfr被计算为计划得分低于90%的计划与最佳计划得分为100%的计划的百分比。结果：FTDC与AXB在规划靶体积（PTV）覆盖率方面表现出极好的一致性，PTV D95% DCEFTDC平均为0.8%±0.9%，而DCEMRDC为-2.5%±3.2%。与MRDC相比，FTDC优化胸椎桨的dce减少且变化较小。FTDC的PFR为10%（50人中有5人），MRDC为32%（50人中有16人）。FTDC的再优化率为4%，而MRDC的再优化率为38%。与没有GPU加速的MRDC相比，带有GPU加速的FTDC平均减少了75%的优化时间。结论：与MRDC相比，FTDC的剂量计算精度更高。它在优化过程中的使用提高了规划质量和效率，辅以gpu。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Applied Clinical Medical Physics 医学-核医学

CiteScore

3.60

自引率

19.00%

发文量

331

审稿时长

3 months

期刊介绍： Journal of Applied Clinical Medical Physics is an international Open Access publication dedicated to clinical medical physics. JACMP welcomes original contributions dealing with all aspects of medical physics from scientists working in the clinical medical physics around the world. JACMP accepts only online submission. JACMP will publish: -Original Contributions: Peer-reviewed, investigations that represent new and significant contributions to the field. Recommended word count: up to 7500. -Review Articles: Reviews of major areas or sub-areas in the field of clinical medical physics. These articles may be of any length and are peer reviewed. -Technical Notes: These should be no longer than 3000 words, including key references. -Letters to the Editor: Comments on papers published in JACMP or on any other matters of interest to clinical medical physics. These should not be more than 1250 (including the literature) and their publication is only based on the decision of the editor, who occasionally asks experts on the merit of the contents. -Book Reviews: The editorial office solicits Book Reviews. -Announcements of Forthcoming Meetings: The Editor may provide notice of forthcoming meetings, course offerings, and other events relevant to clinical medical physics. -Parallel Opposed Editorial: We welcome topics relevant to clinical practice and medical physics profession. The contents can be controversial debate or opposed aspects of an issue. One author argues for the position and the other against. Each side of the debate contains an opening statement up to 800 words, followed by a rebuttal up to 500 words. Readers interested in participating in this series should contact the moderator with a proposed title and a short description of the topic