OpenKBP-Opt: an international and reproducible evaluation of 76 knowledge-based planning pipelines.
OpenKBP-Opt: an international and reproducible evaluation of 76 knowledge-based planning pipelines.
复制标题
DOI：
10.1088/1361-6560/ac8044
复制
发表时间：
2022-09-12
影响因子：
3.5
通讯作者：
Chan TCY
中科院分区：
文献类型：
--
作者：
Babier A;Mahmood R;Zhang B;Alves VGL;Barragán-Montero AM;Beaudry J;Cardenas CE;Chang Y;Chen Z;Chun J;Diaz K;David Eraso H;Faustmann E;Gaj S;Gay S;Gronberg M;Guo B;He J;Heilemann G;Hira S;Huang Y;Ji F;Jiang D;Carlo Jimenez Giraldo J;Lee H;Lian J;Liu S;Liu KC;Marrugo J;Miki K;Nakamura K;Netherton T;Nguyen D;Nourzadeh H;Osman AFI;Peng Z;Darío Quinto Muñoz J;Ramsl C;Joo Rhee D;David Rodriguez J;Shan H;Siebers JV;Soomro MH;Sun K;Usuga Hoyos A;Valderrama C;Verbeek R;Wang E;Willems S;Wu Q;Xu X;Yang S;Yuan L;Zhu S;Zimmermann L;Moore KL;Purdie TG;McNiven AL;Chan TCY
建立一个开放框架来开发基于知识的规划（KBP）的规划优化模型。我们的框架包括 100 名接受调强放射治疗的头颈癌患者的放射治疗数据（即参考计划）。这些数据还包括 19 个 KBP 模型的高质量剂量预测，这些模型是由不同研究小组在 OpenKBP Grand Challenge 期间使用样本外数据开发的。将剂量预测输入到四个基于注量的剂量模拟模型中，形成 76 个独特的 KBP 管道，生成 7600 个计划（76 个管道×100 名患者）。通过以下方式将预测和 KBP 生成的计划与参考计划进行比较： 剂量评分，即剂量的平均绝对体素差异；剂量体积直方图 (DVH) 点的偏差；以及临床计划标准满足的频率。我们还进行了理论研究来证明我们的剂量模拟模型的合理性。预测与其 KBP 管道之间的剂量评分的排序相关性范围为 0.50-0.62，这表明预测的质量通常与计划的质量正相关。此外，与输入预测相比，KBP 生成的计划表现明显更好，P<0.05；单侧 Wilcoxon 检验）对 23 个 DVH 点中的 18 个点进行测试。同样，每个优化模型生成的计划比参考计划满足更高百分比的标准，比所有剂量预测集多满足 3.5% 的标准。最后，我们的理论研究表明，剂量模拟模型生成的计划对于逆向计划模型也是最佳的。这是迄今为止评估 KBP 预测和优化模型组合的最大国际努力。我们发现表现最好的模型显着优于参考剂量和剂量预测。为了可重复性，我们的数据和代码是免费提供的。
To establish an open framework for developing plan optimization models for knowledge-based planning (KBP). Our framework includes radiotherapy treatment data (i.e. reference plans) for 100 patients with head-and-neck cancer who were treated with intensity-modulated radiotherapy. That data also includes high-quality dose predictions from 19 KBP models that were developed by different research groups using out-of-sample data during the OpenKBP Grand Challenge. The dose predictions were input to four fluence-based dose mimicking models to form 76 unique KBP pipelines that generated 7600 plans (76 pipelines×100 patients). The predictions and KBP-generated plans were compared to the reference plans via: the dose score, which is the average mean absolute voxel-by-voxel difference in dose; the deviation in dose-volume histogram (DVH) points; and the frequency of clinical planning criteria satisfaction. We also performed a theoretical investigation to justify our dose mimicking models. The range in rank order correlation of the dose score between predictions and their KBP pipelines was 0.50–0.62, which indicates that the quality of the predictions was generally positively correlated with the quality of the plans. Additionally, compared to the input predictions, the KBP-generated plans performed significantly better P<0.05; one-sided Wilcoxon test) on 18 of 23 DVH points. Similarly, each optimization model generated plans that satisfied a higher percentage of criteria than the reference plans, which satisfied 3.5% more criteria than the set of all dose predictions. Lastly, our theoretical investigation demonstrated that the dose mimicking models generated plans that are also optimal for an inverse planning model. This was the largest international effort to date for evaluating the combination of KBP prediction and optimization models. We found that the best performing models significantly outperformed the reference dose and dose predictions. In the interest of reproducibility, our data and code is freely available.
登录
查看更多内容
影响因子：
3.8
作者：
Babier, Aaron;Zhang, Binghao;Chan, Timothy C. Y.
通讯作者：
Chan, Timothy C. Y.
影响因子：
1.2
作者：
Breedveld S;Heijmen B
通讯作者：
Heijmen B
影响因子：
3.8
作者：
Gronberg, Mary P.;Gay, Skylar S.;Cardenas, Carlos E.
通讯作者：
Cardenas, Carlos E.
影响因子：
3.8
作者：
Deasy, JO;Blanco, AI;Clark, VH
通讯作者：
Clark, VH
影响因子：
3.6
作者：
Bai, Penggang;Weng, Xing;Chen, Chuanben
通讯作者：
Chen, Chuanben