An ultra-fast deep-learning-based dose engine for prostate VMAT via knowledge distillation framework with limited patient data
An ultra-fast deep-learning-based dose engine for prostate VMAT via knowledge distillation framework with limited patient data
复制标题
基于有限患者数据的知识蒸馏框架，基于超快速深度学习的前列腺 VMAT 剂量引擎
DOI：
10.1088/1361-6560/aca5eb
复制
发表时间：
2022
影响因子：
3.5
通讯作者：
Lu, Bo
中科院分区：
文献类型：
--
作者：
Tseng, Wenchih;Liu, Hongcheng;Yang, Yu;Liu, Chihray;Lu, Bo
目的为解决传统剂量计算算法计算精度与效率之间的内在矛盾，开发了基于深度学习的剂量引擎。然而，当前基于dl的引擎通常具有很高的计算复杂度，并且需要强大的计算设备。因此，为了减轻其计算负担并扩大其在资源有限设备的临床环境中的适用性，我们提出了一个通过知识蒸馏（KD）框架的紧凑剂量引擎，该框架为前列腺体积调制弧线治疗（VMAT）提供了超快的计算速度和高精度。KD框架包含两个子模型：一个大的预培训教师和一个小的待培训学生。学生接受老师传授的知识，以便更好地归纳。经过训练的学生充当剂量计算的最终引擎。模型输入为患者计算机断层扫描和水中VMAT剂量，输出为dl计算的患者剂量。地基真值剂量由摩纳哥治疗计划系统的蒙特卡罗模块计算。分别纳入20例和10例前列腺病例进行模型训练和评估。通过Gamma分析和推理效率来评估模型的性能（教师/学生/仅学生）。剂量学比较（输入剂量/ dl计算剂量/地面真值剂量）表明，所提出的发动机可以有效地将水中低精度剂量转换为高精度患者剂量。dl计算剂量与地面真实剂量之间的Gamma通过率（2%/2 mm， 10%阈值）为98.64±0.62%（教师），98.13±0.76%（学生）和96.95±1.02%（学生）。使用图形处理单元设备的推理时间为16毫秒（教师）和11毫秒（学生/学生专用），而使用中央处理单元设备的推理时间为936毫秒（教师）和374毫秒（学生/学生专用）。在KD框架下，小型剂量机可以达到与大型剂量机相当的精度。其体积小巧，减少了计算量和对计算设备的要求，更具有临床应用价值。
ObjectiveDeep-learning (DL)-based dose engines have been developed to alleviate the intrinsic compromise between the calculation accuracy and efficiency of the traditional dose calculation algorithms. However, current DL-based engines typically possess high computational complexity and require powerful computing devices. Therefore, to mitigate their computational burdens and broaden their applicability to a clinical setting where resource-limited devices are available, we proposed a compact dose engine via knowledge distillation (KD) framework that offers an ultra-fast calculation speed with high accuracy for prostate Volumetric Modulated Arc Therapy (VMAT).ApproachThe KD framework contains two sub-models: a large pre-trained teacher and a small to-be-trained student. The student receives knowledge transferred from the teacher for better generalization. The trained student serves as the final engine for dose calculation. The model input is patient computed tomography and VMAT dose in water, and the output is DL-calculated patient dose. The ground-truth\dose was computed by the Monte Carlo module of the Monaco treatment planning system. Twenty and ten prostate cases were included for model training and assessment, respectively. The model's performance (teacher/student/student-only) was evaluated by Gamma analysis and inference efficiency.Main resultsThe dosimetric comparisons (input/DL-calculated/ground-truth doses) suggest that the proposed engine can effectively convert low-accuracy doses in water to high-accuracy patient doses. The Gamma passing rate (2%/2 mm, 10% threshold) between the DL-calculated and ground-truth doses was 98.64±0.62%(teacher), 98.13±0.76%(student), and 96.95±1.02%(student-only). The inference time was 16 milliseconds (teacher) and 11 milliseconds (student/student-only) using a graphics processing unit device, while it was 936 milliseconds (teacher) and 374 milliseconds (student/student-only) using a central processing unit device.SignificanceWith the KD framework, a compact dose engine can achieve comparable accuracy to that of a larger one. Its compact size reduces the computational burdens and computing device requirements, and thus such an engine can be more clinically applicable.
DOI：
10.1016/j.meddos.2013.10.001
发表时间：
2014
期刊：
Medical dosimetry : official journal of the American Association of Medical Dosimetrists
影响因子：
--
作者：
Chen H;Craft DL;Gierga DP
通讯作者：
Gierga DP
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：
--
发表时间：
2021
期刊：
Medical Physics (Lancaster)
影响因子：
--
作者：
Cong Liu;Xinye Ni;Xiance Jin;Wen Si
通讯作者：
Wen Si