Automatic plan selection using deep network—A prostate study

IF 3.2 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Medical physics Pub Date : 2024-12-10 DOI:10.1002/mp.17550

Philippe Y. Chatigny, Cédric Bélanger, Éric Poulin, Luc Beaulieu

{"title":"Automatic plan selection using deep network—A prostate study","authors":"Philippe Y. Chatigny, Cédric Bélanger, Éric Poulin, Luc Beaulieu","doi":"10.1002/mp.17550","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Background</h3>\n \n <p>Recently, high-dose-rate (HDR) brachytherapy treatment plans generation was improved with the development of multicriteria optimization (MCO) algorithms that can generate thousands of pareto optimal plans within seconds. This brings a shift, from the objective of generating an acceptable plan to choosing the best plans out of thousands.</p>\n </section>\n \n <section>\n \n <h3> Purpose</h3>\n \n <p>In order to choose the best plans, new criteria beyond usual dosimetrics volumes histogram (DVH) metrics are introduced and a deep learning (DL) framework is added as an automatic plan selection algorithm.</p>\n </section>\n \n <section>\n \n <h3> Methods</h3>\n \n <p>The new criteria are visual-like criteria implemented for the bladder, rectum, and urethra. One criterion also takes into account the cold spot in the prostate. Those criteria, along with commonly used DVH criteria, are used to form classes on which to train the algorithm. The algorithm is trained with an input of two 3D images, dose and mask of the anatomy, in order to rank and automatically select a plan. The confidence in the output is used for ranking and the automatic plan selection. The algorithm is trained on 835 previously treated prostate cancer patients and evaluated on a separated 20 patients cohort previously evaluated by two experts (clinical medical physicists) in an inter-observer MCO study.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>The deep network takes 10 s to rank 2000 plans (vs. 5–10 min for experts to rank 4 preferred plans). A total of four different networks are trained which offer different trade-offs. The key trade-offs are the target coverage or the organs at risk (OAR) sparing. The algorithm with the best network achieves no statistical difference with the plans chosen by the two experts for 6 and 9 criteria, respectively, out of 13 criteria (paired <i>t</i>-test with <i>p</i> <span></span><math>\n <semantics>\n <mo>></mo>\n <annotation>$>$</annotation>\n </semantics></math> 0.05) while the two experts have no statistical difference between them for 7 criteria.</p>\n </section>\n \n <section>\n \n <h3> Conclusions</h3>\n \n <p>The developed approach is flexible since it allows the modification or addition of criteria to obtain different trade-offs in plan quality, per the institution standard. The approach is fast and robust while adding negligible time to MCO planning. These results demonstrate potential for clinical use.</p>\n </section>\n </div>","PeriodicalId":18384,"journal":{"name":"Medical physics","volume":"52 3","pages":"1717-1727"},"PeriodicalIF":3.2000,"publicationDate":"2024-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/mp.17550","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical physics","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/mp.17550","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Recently, high-dose-rate (HDR) brachytherapy treatment plans generation was improved with the development of multicriteria optimization (MCO) algorithms that can generate thousands of pareto optimal plans within seconds. This brings a shift, from the objective of generating an acceptable plan to choosing the best plans out of thousands.

Purpose

In order to choose the best plans, new criteria beyond usual dosimetrics volumes histogram (DVH) metrics are introduced and a deep learning (DL) framework is added as an automatic plan selection algorithm.

Methods

The new criteria are visual-like criteria implemented for the bladder, rectum, and urethra. One criterion also takes into account the cold spot in the prostate. Those criteria, along with commonly used DVH criteria, are used to form classes on which to train the algorithm. The algorithm is trained with an input of two 3D images, dose and mask of the anatomy, in order to rank and automatically select a plan. The confidence in the output is used for ranking and the automatic plan selection. The algorithm is trained on 835 previously treated prostate cancer patients and evaluated on a separated 20 patients cohort previously evaluated by two experts (clinical medical physicists) in an inter-observer MCO study.

Results

The deep network takes 10 s to rank 2000 plans (vs. 5–10 min for experts to rank 4 preferred plans). A total of four different networks are trained which offer different trade-offs. The key trade-offs are the target coverage or the organs at risk (OAR) sparing. The algorithm with the best network achieves no statistical difference with the plans chosen by the two experts for 6 and 9 criteria, respectively, out of 13 criteria (paired t-test with p $>$ 0.05) while the two experts have no statistical difference between them for 7 criteria.

Conclusions

The developed approach is flexible since it allows the modification or addition of criteria to obtain different trade-offs in plan quality, per the institution standard. The approach is fast and robust while adding negligible time to MCO planning. These results demonstrate potential for clinical use.

Abstract Image

查看原文本刊更多论文

利用深度网络自动选择计划--前列腺研究。

背景：最近，随着多标准优化（MCO）算法的发展，高剂量率（HDR）近距离放射治疗计划的生成得到了改进，该算法可在数秒内生成数千个帕累托最优计划。目的：为了选择最佳方案，除了通常的剂量体积直方图（DVH）指标外，还引入了新的标准，并添加了深度学习（DL）框架作为自动方案选择算法：新标准是针对膀胱、直肠和尿道实施的视觉类标准。其中一个标准还考虑到了前列腺的冷点。这些标准与常用的 DVH 标准一起构成了对算法进行训练的类别。该算法使用两张三维图像、剂量和解剖掩膜作为输入进行训练，以便对计划进行排序和自动选择。输出的置信度用于排序和自动选择方案。该算法在 835 名先前接受过治疗的前列腺癌患者身上进行了训练，并在一项观察者间 MCO 研究中，在先前由两名专家（临床医学物理学家）评估过的 20 名患者队列中进行了评估：深度网络只需 10 秒钟就能对 2000 个方案进行排序（相比之下，专家需要 5-10 分钟才能对 4 个首选方案进行排序）。共训练了四个不同的网络，它们提供了不同的权衡。关键的权衡因素是目标覆盖范围或风险器官（OAR）疏通。在 13 项标准中，最佳网络算法分别在 6 项和 9 项标准上与两位专家选择的计划没有统计学差异（配对 t 检验，p > $>$ 0.05），而两位专家在 7 项标准上没有统计学差异：所开发的方法非常灵活，因为它允许修改或添加标准，以便根据机构标准对计划质量进行不同的权衡。该方法既快速又稳健，同时给 MCO 规划增加的时间可以忽略不计。这些结果证明了其临床应用的潜力。

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

Medical physics 医学-核医学

CiteScore

6.80

自引率

15.80%

发文量

660

审稿时长

1.7 months

期刊介绍： Medical Physics publishes original, high impact physics, imaging science, and engineering research that advances patient diagnosis and therapy through contributions in 1) Basic science developments with high potential for clinical translation 2) Clinical applications of cutting edge engineering and physics innovations 3) Broadly applicable and innovative clinical physics developments Medical Physics is a journal of global scope and reach. By publishing in Medical Physics your research will reach an international, multidisciplinary audience including practicing medical physicists as well as physics- and engineering based translational scientists. We work closely with authors of promising articles to improve their quality.