Lv Jun, Chen Liuli, Zhu Zhiqiang, Long Pengcheng, Hu Liqin, Zhou Han, Shen Zetian
Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, China.
University of Science and Technology of China, Hefei, China.
Med Phys. 2025 May;52(5):3439-3449. doi: 10.1002/mp.17690. Epub 2025 Feb 18.
Multi-leaf collimators (MLCs) are crucial for modern radiotherapy as they ensure precise target irradiation through accurate leaf positioning. Accurate prediction of MLC leaf positions is vital for the effectiveness and safety of treatments.
This study aims to establish three neural network models for predicting the delivered positions of MLCs in radiotherapy.
Fifty plans with sliding window dynamic intensity-modulated radiation therapy delivery were selected from an Elekta linear accelerator, which features a 160-leaf MLC system. The dose fraction, gantry angle, collimator angle, X1 and X2 jaw positions, Y1 and Y2 carriage positions, planned leaf positions, adjacent leaf positions, leaf gap, leaf velocity, and leaf acceleration were extracted from the planning data in the machine's log files and used as model inputs, with the delivered leaf positional serving as the target response. This establishes the input-output relationship for the neural network, and the predicted MLC positions are obtained through training. Particle Swarm Optimization Back Propagation Neural Network (PSOBPNN), Back Propagation Neural Network (BPNN), and Radial Basis Function Neural Network (RBFNN) architectures were developed to predict MLC leaf positional deviations during treatment. The training was conducted on 70% of the sample data, with the remaining 30% used for validation and testing. Model performance was assessed using metrics such as mean absolute error (MAE), mean squared error (MSE), regression plots, and error histograms.
The proposed neural network models demonstrated high accuracy in predicting MLC leaf positions. The PSOBPNN model demonstrated superior performance with an MAE of 0.0043 mm and an MSE of 0.00003 mm. In comparison, the BPNN model achieved an MAE of 0.0241 mm and an MSE of 0.001 mm, while the RBFNN model exhibited an MAE of 0.0331 mm and an MSE of 0.0019 mm. The correlation coefficient (R = 0.9999) of models indicates a close match between predicted and delivered leaf positions for all MLC leaves.
Three models were evaluated for predicting the delivered MLC positions using data from an Elekta accelerator. The PSOBPNN model exhibited superior performance by achieving markedly lower MAE and MSE values while also demonstrating robust generalizability in predicting positions across various leaf indices, outperforming the conventional BPNN and RBFNN models.
多叶准直器（MLC）对现代放射治疗至关重要，因为它们通过精确的叶片定位确保对靶区的精确照射。准确预测MLC叶片位置对于治疗的有效性和安全性至关重要。
本研究旨在建立三个神经网络模型，用于预测放射治疗中MLC的实际输送位置。
从配备160叶MLC系统的医科达直线加速器中选择50个采用滑动窗口动态调强放射治疗输送的计划。从机器日志文件中的计划数据中提取剂量分割、机架角度、准直器角度、X1和X2光阑位置、Y1和Y2托架位置、计划叶片位置、相邻叶片位置、叶片间隙、叶片速度和叶片加速度，并将其用作模型输入，以实际输送的叶片位置作为目标响应。这建立了神经网络的输入输出关系，并通过训练获得预测的MLC位置。开发了粒子群优化反向传播神经网络（PSOBPNN）、反向传播神经网络（BPNN）和径向基函数神经网络（RBFNN）架构，以预测治疗期间MLC叶片位置偏差。使用70%的样本数据进行训练，其余30%用于验证和测试。使用平均绝对误差（MAE）、均方误差（MSE）、回归图和误差直方图等指标评估模型性能。
所提出的神经网络模型在预测MLC叶片位置方面表现出高精度。PSOBPNN模型表现出卓越性能，MAE为0.0043毫米，MSE为0.00003毫米。相比之下，BPNN模型的MAE为0.0241毫米，MSE为0.001毫米，而RBFNN模型的MAE为0.0331毫米，MSE为0.0019毫米。模型的相关系数（R = 0.9999）表明，所有MLC叶片的预测位置与实际输送位置密切匹配。
使用医科达加速器的数据对三个预测MLC实际输送位置的模型进行了评估。PSOBPNN模型表现出卓越性能，MAE和MSE值显著更低，同时在预测不同叶片索引的位置时也表现出强大的泛化能力，优于传统的BPNN和RBFNN模型。
