NeuralDAO: Incorporating neural-network-generated dose into direct aperture optimization for end-to-end imrt planning.
NeuralDAO: Incorporating neural-network-generated dose into direct aperture optimization for end-to-end imrt planning.
复制标题
NeuralDAO：将神经网络生成的剂量纳入直接孔径优化中，以实现端到端 IMRT 规划。
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Wen Si
中科院分区：
文献类型：
--
作者：
Cong Liu;Xinye Ni;Xiance Jin;Wen Si
目的
调强放射治疗（IMRT）计划中的当前实践几乎总是包括用于计划优化和最终剂量验证的不同剂量计算策略。精确的蒙特卡罗（MC）剂量算法被认为是耗时的优化。因此，在优化中使用快速、简化的剂量算法。优化剂量和输送剂量之间的显著差异导致繁琐的规划循环和潜在的次优解决方案。本工作旨在开发一种调强放射治疗优化算法，以最大限度地减少剂量差异，使输送剂量可以在一个整体，端到端的方式进行优化。
方法
所提出的算法，即NeuralDAO，集成了神经剂量网络的列生成（CG）直接孔径优化（DAO）制定的步进和拍摄调强放射治疗计划。神经剂量网络的设计和训练，以产生剂量的MC级精度在几毫秒内。它的可微性被充分利用来计算用于识别潜在孔径形状的梯度。NeuralDAO的原型是在PyTorch中开发的，并向公众开放。研究了5例肺部患者病例。将剂量测定准确度与MC剂量进行比较。将计划质量和时间与最先进的（SoA）剂量校正算法进行比较。通过Wilcoxon符号秩检验进行统计分析。
结果
在优化剂量和输送剂量之间，2mm2%处的平均伽马通过率为99.7%。NeuralDAO产生的收敛过程实际上与基于MC的DAO产生的收敛过程相同。对于GPU上的孔径，平均剂量计算时间为12.1毫秒。一次优化需要10分钟到36分钟。与SoA相比，观察到靶的符合性指数和均匀性指数更好。食道明显幸免。观察到重新规划的数量和规划时间大幅减少。
结论
提出了一种基于神经剂量网络的DAO算法。结果表明，该算法最大限度地减少了优化和交付剂量之间的差异，这提供了一个有前途的方法，以减少在调强放射治疗计划所需的时间和精力。这项工作证明了神经网络在调强放射治疗优化中应用的可能性。这是一个很大的潜力，将该算法扩展到其他治疗方式。本文受版权保护。All rights reserved.
PURPOSE
The current practice in Intensity Modulated Radiation Therapy (IMRT) planning almost always includes different dose calculation strategies for plan optimization and final dose verification. The accurate Monte Carlo (MC) dose algorithm is considered to be time-consuming for the optimization. Thus a fast, simplified dose algorithm is used in the optimization. The significant differences between the optimized dose and the delivered dose lead to tediously-planning loops and potentially-suboptimal solutions. This work aims to develop an IMRT optimization algorithm to minimize the dose discrepancy so that the delivered dose can be optimized in a holistic, end-to-end manner.
METHODS
The proposed algorithm, namely NeuralDAO, integrates a neural dose network into the Column-Generation (CG) Direct Aperture Optimization (DAO) formulation for step-and-shoot IMRT planning. The neural dose network is designed and trained to produce doses of MC-level accuracy within few milliseconds. Its differentiability is fully exploited to compute gradients for identifying potential aperture shapes. A prototype of NeuralDAO was developed in PyTorch and available to the public. Five lung patient cases have been studied. Dosimetric accuracy was compared with the MC dose. Plan quality and time were compared with a state-of-the-art (SoA) dose-correct algorithm. Statistical analysis was performed by Wilcoxon signed-rank test.
RESULTS
The average gamma passing rate at 2mm2% is 99.7% between the optimized and delivered dose. The Convergence process produced by NeuralDAO is virtually identical to that produced by an MC-based DAO. The average dose calculation time is 12.1 milliseconds for an aperture on GPU. One session of optimization took 10 minutes to 36 minutes. Compared with the SoA, a better conformity index and homogeneity index were observed for the target. The esophagus was significantly spared. Significant reductions were observed for the re-planning number and the planning time.
CONCLUSIONS
A new DAO algorithm based on the neural dose network has been developed. The results suggest this algorithm minimizes the discrepancy between the optimized and delivered dose, which offers a promising approach to reduce the time and effort required in IMRT planning. This work demonstrates the possibility of applying the neural network in IMRT optimization. It is of great potential to extend this algorithm to other treatment modalities. This article is protected by copyright. All rights reserved.
影响因子：
3.8
作者：
J. Siebers;I. Kawrakow;Viswanathan Ramakrishnan
通讯作者：
J. Siebers;I. Kawrakow;Viswanathan Ramakrishnan
在肺等效介质中对 DPM 蒙特卡罗代码进行光子束相对剂量验证。
影响因子：
3.8
作者：
Chetty,IndrinJ;Charland,PauleM;Tyagi,Neelam;McShan,DanielL;Fraass,BenedickA;Bielajew,AlexF
通讯作者：
Bielajew,AlexF
影响因子：
3.8
作者：
Siebers, JV;Tong, SD;Mohan, R
通讯作者：
Mohan, R