Evaluating machine learning enhanced intelligent-optimization-engine (IOE) performance for ethos head-and-neck (HN) plan generation.
Evaluating machine learning enhanced intelligent-optimization-engine (IOE) performance for ethos head-and-neck (HN) plan generation.
复制标题
评估机器学习增强了精神主颈（HN）计划生成的智能优化引擎（IOE）性能。
DOI：
10.1002/acm2.13950
复制
发表时间：
2023-07
影响因子：
2.1
通讯作者：
Lin, Mu-Han
中科院分区：
文献类型：
--
作者：
Visak, Justin;Inam, Enobong;Meng, Boyu;Wang, Siqiu;Parsons, David;Nyugen, Dan;Zhang, Tingliang;Moon, Dominic;Avkshtol, Vladimir;Jiang, Steve;Sher, David;Lin, Mu-Han
Varian Ethos 利用新颖的智能优化引擎 (IOE) 来实现规划自动化。然而，这引入了一种黑盒方法来优化计划，并为规划者提高计划质量带来了挑战。本研究旨在评估机器学习引导的头颈 (H&N) 适应性放射治疗 (ART) 初始参考计划生成方法。使用固定的 18 束调强放射治疗 (IMRT) 模板，在 Ethos 计划系统中对 20 名既往接受过 C 形臂/环安装治疗的患者进行了回顾性重新计划。 IOE 输入的临床目标是使用（1）内部深度学习 3D 剂量预测器（AI 引导）（2）基于商业知识的规划（KBP）模型和基于通用 RTOG 的人群标准（KBP-RTOG）和（3）仅基于 RTOG 的约束模板（RTOG）生成的，用于深入分析 IOE 敏感性。两个模型都使用了类似的训练数据。对计划进行优化，直到达到各自的标准或满足 DVH 估计范围。计划已标准化，最高 PTV 剂量水平达到 95% 的覆盖率。与临床（基准）计划相比，评估了目标覆盖率、高影响器官风险（OAR）和计划的可实施性。使用配对双尾学生 t 检验评估统计显着性。就临床基准案例而言，人工智能引导的计划优于 KBP-RTOG 和仅 RTOG 计划。总体而言，与基准相比，AI 指导计划的 OAR 剂量相当或有所改善，而 KBP-RTOG 和 RTOG 计划则增加了 OAR 剂量。然而，所有计划总体上都满足 RTOG 标准。所有计划的异质性指数 (HI) 平均 <1.07。 KBP-RTOG、AI-Guided、RTOG 和基准计划的平均调制因子分别为 12.2 ± 1.9 (p = n.s.)、13.1 ± 1.4 (p = <0.001)、11.5 ± 1.3 (p = n.s.) 和 12.2 ± 1.9。人工智能引导的计划质量最高。随着诊所采用 ART 工作流程，支持 KBP 的计划和仅支持 RTOG 的计划都是可行的方法。与约束优化类似，IOE 对临床输入目标很敏感，我们建议与机构的规划指令剂量测定标准相当的输入。
Varian Ethos utilizes novel intelligent‐optimization‐engine (IOE) designed to automate the planning. However, this introduced a black box approach to plan optimization and challenge for planners to improve plan quality. This study aims to evaluate machine‐learning‐guided initial reference plan generation approaches for head & neck (H&N) adaptive radiotherapy (ART). Twenty previously treated patients treated on C‐arm/Ring‐mounted were retroactively re‐planned in the Ethos planning system using a fixed 18‐beam intensity‐modulated radiotherapy (IMRT) template. Clinical goals for IOE input were generated using (1) in‐house deep‐learning 3D‐dose predictor (AI‐Guided) (2) commercial knowledge‐based planning (KBP) model with universal RTOG‐based population criteria (KBP‐RTOG) and (3) an RTOG‐based constraint template only (RTOG) for in‐depth analysis of IOE sensitivity. Similar training data was utilized for both models. Plans were optimized until their respective criteria were achieved or DVH‐estimation band was satisfied. Plans were normalized such that the highest PTV dose level received 95% coverage. Target coverage, high‐impact organs‐at‐risk (OAR) and plan deliverability was assessed in comparison to clinical (benchmark) plans. Statistical significance was evaluated using a paired two‐tailed student t‐test. AI‐guided plans were superior to both KBP‐RTOG and RTOG‐only plans with respect to clinical benchmark cases. Overall, OAR doses were comparable or improved with AI‐guided plans versus benchmark, while they increased with KBP‐RTOG and RTOG plans. However, all plans generally satisfied the RTOG criteria. Heterogeneity Index (HI) was on average <1.07 for all plans. Average modulation factor was 12.2 ± 1.9 (p = n.s), 13.1 ± 1.4 (p = <0.001), 11.5 ± 1.3 (p = n.s.) and 12.2 ± 1.9 for KBP‐RTOG, AI‐Guided, RTOG and benchmark plans, respectively. AI‐guided plans were the highest quality. Both KBP‐enabled and RTOG‐only plans are feasible approaches as clinics adopt ART workflows. Similar to constrained optimization, the IOE is sensitive to clinical input goals and we recommend comparable input to an institution's planning directive dosimetric criteria.
登录
查看更多内容
影响因子：
2.1
作者：
Moazzezi M;Rose B;Kisling K;Moore KL;Ray X
通讯作者：
Ray X
影响因子：
3.1
作者：
Sher DJ;Godley A;Park Y;Carpenter C;Nash M;Hesami H;Zhong X;Lin MH
通讯作者：
Lin MH
影响因子：
3.8
作者：
Barragan-Montero, Ana Maria;Dan Nguyen;Jiang, Steve
通讯作者：
Jiang, Steve
影响因子：
3.6
作者：
Holt, Andrea;Van Gestel, Dirk;van Vliet-Vroegindeweij, Corine
通讯作者：
van Vliet-Vroegindeweij, Corine
影响因子：
2.1
作者：
Byrne M;Archibald-Heeren B;Hu Y;Teh A;Beserminji R;Cai E;Liu G;Yates A;Rijken J;Collett N;Aland T
通讯作者：
Aland T