基于GPU加速的质子调强放疗鲁棒优化器

开发一种基于图形处理器（GPU）加速的质子调强放疗鲁棒优化器,用于减小质子束射程不确定性和靶区定位偏差对质子放疗的影响。建立的鲁棒优化模型使用的目标函数包括9种边界剂量目标,分别是:无偏差情况、2种射程偏差（偏长与偏短）、6种摆位不确定性（前后、侧向、上下入射方向各2种正负偏差）。首先靶区和危及器官的剂量贡献矩阵使用笔形束算法计算得到,然后使用共轭梯度法优化目标函数让其满足约束条件,这两部分均采用GPU加速。头颈部、肺部和前列腺三个临床病例被用来检测本优化器的性能表现。与传统基于计划靶区（PTV）的质子调强放疗计划相比,鲁棒优化器能够优化出对射程不确定性和摆位误差更加不敏感的治疗计划,让靶区实现了高剂量均匀性的同时危及器官（OARs）也得到了更好的保护。经过100次迭代,三个病例的优化时间均在10 s左右。该结果证明了基于GPU加速的质子调强放疗鲁棒优化器能够在短时间内设计出高鲁棒性的质子治疗计划,从而提高质子放射治疗的可靠性。This paper describes the development of a fast robust optimization tool that takes advantage of the GPU technologies. The objective function of the robust optimization model considered nine boundary dose distributions--two for ±range uncertainties, six for ±set-up uncertainties along anteroposterior (A-P), lateral (R-L) and superior{inferior (S-I) directions, and one for nominal situation. The nine boundary influence matrices were calculated using an in-house dose engine for proton pencil beams of a finite size, while the conjugate gradient method was applied to minimize the objective function. The GPU platform was adopted to accelerate both the proton dose calculation algorithm and the conjugate gradient method. Three clinical cases-one head and neck cancer case, one lung cancer case and one prostate cancer case-were investigated to demonstrate the clinical significance of the proposed robust optimizer. Compared with conventional planning target volume (PTV) based IMPT plans, the proposed method was found to be conducive in designing robust treatment plans that were less sensitive to range and setup uncertainties. The three cases showed that targets could achieve high dose uniformity while organs at risks (OARs) were under better protection against setup and range errors. The run times for the three cases were around 10 s for 100 iterations. The GPU-based fast robust optimizer developed in this study can serve to improve the reliability of traditional proton treatment planning by achieving a high level of robustness in a much shorter time.     

基于角度优化的鲁棒极端学习机算法

针对数据集样本中带有噪声和离群点问题,提出了一种基于角度优化的鲁棒极端学习机算法。该方法利用鲁棒激活函数角度优化的原则,首先降低了离群点对分类算法的影响,从而保持数据样本的全局结构信息,达到更好的去噪效果。其次,有效的避免隐层节点输出矩阵求解不准的问题,进一步增强极端学习机的泛化性能。通过应用在普遍图像数据库上的实验结果表明,这种提出的算法与其他算法相比具有更强的鲁棒性和较高的识别率。     

大分割碳离子放射治疗中相对生物学效应与剂量依赖关系的研究

应用碳离子束进行大分割放射治疗从而缩短治疗周期是碳离子束治疗的优势之一。为研究大分割放疗增加单次照射剂量后,碳离子束相对生物学效应（RBE）的变化,应用细胞存活线性平方（LQ）模型推导出RBE与剂量的依赖关系。基于此关系研究了具有不同辐射敏感性的肿瘤细胞和正常组织细胞RBE随剂量的变化。结果表明,在0~20 Gy范围内,不论肿瘤细胞与正常组织细胞具有怎样的辐射敏感性,肿瘤细胞的RBE值始终大于正常组织细胞。此外,基于理论推导和对相关实验数据的分析,证实了RBE随剂量增加而递增现象的存在。这些结果对应用碳离子束进行大分割放射治疗具有重要的指导作用。Short treatment course due to the suitability of hypofractionated regimen for carbon ion beam is one of the advantages of carbon ion radiotherapy. To study the dependence of relative biological effectiveness (RBE) on dose, the relationship between RBE and dose was deduced through the linear-quadratic (LQ) model. Based on the relationship, the change of RBE of tumor cells and normal tissue cells with different radiosensitivities with dose was studied. The results showed that the RBE value of tumor cells was always greater than that of normal tissue cells in the dose range of 0~20 Gy, regardless of the radiosensitivity of tumor and normal tissue cells. In addition, based on theoretical deduction and analysis of the relevant experimental data, the existence of RBE increase with increasing dose was verified. These results are of great significance for conducting hypofractionated radiotherapy with carbon ion beam.     

碳离子束混合LET辐照的EBT3辐射变色胶片剂量修正方法

针对EBT3辐射变色胶片对碳离子束混合LET辐照的剂量欠响应比较了两种剂量修正方法。利用260 MeV/u的碳离子束通过被动降能得到多种剂量平均LET的碳离子束,利用这些碳离子束进行了胶片剂量响应刻度辐照,选择最佳的拟合公式得到了胶片剂量刻度曲线。使用RE(Relative Efficiency)量化了EBT3胶片随LET的剂量欠响应,并使用RE剂量修正法修正了混合LET辐照胶片的剂量。此外,根据剂量刻度曲线公式中拟合参数随不同LET所占剂量比例的变化规律,提出了拟合参数剂量修正法并修正了混合LET辐照胶片的剂量。最后比较了这两种方法的结果,表明拟合参数方法得到的剂量偏差在5%以内,优于RE方法10%以内的剂量偏差。     

均匀磁场下碳离子笔形束的剂量变化分析及位置修正方法

分析外加均匀磁场对于碳离子笔形束剂量分布的影响,并考虑修正这种影响,为磁共振成像引导碳离子放射治疗的临床应用提供指导。本文利用蒙特卡罗方法模拟计算了不同能量碳离子笔形束在不同强度磁场下的剂量分布情况,发现垂直于碳离子束入射方向的均匀磁场对于碳离子笔形束射程缩短的影响很小,磁场对碳离子束的主要影响是引起束流横向偏转,特别是碳离子束布拉格峰位置的横向侧移。横向侧移程度与碳离子束的能量和磁场强度相关,根据模拟结果,得到了一个计算碳离子束布拉格峰在磁场中相对横向偏转的方程,并提出一种校正外加磁场引起的碳离子束布拉格峰横移的角度修正方法。这些结果可用于指导磁共振图像引导碳离子放射治疗计划系统的研发。     

重离子治疗计划的生物有效剂量及其验证

介绍了重离子治疗计划中的生物有效剂量，以及基于局部效应模型的相对生物学效应的计算. 报告了用生物测试系统测得的实验结果，并与治疗计划的计算结果做了比较. 靶子内外结果的一致性显示了重离子治疗计划的可靠性及保护紧挨着靶的重要器官的能力.     

步进扫描投影光刻机剂量控制参量优化新算法研究

提出了一种以生产率、剂量精度与激光器使用成本三者最佳匹配为优化目标的步进扫描投影光刻机剂量控制参量优化新算法及其数学模型,通过将激光器重复频率作为可调参量并引入有效脉冲个数的概念,获得了有效剂量区间内任意给定剂量所应采取的优化策略并给出了剂量控制参量的具体计算方法。理论推导和模拟计算结果表明,新算法既保持了原算法在生产率和剂量精度优化方面的优势,又改进了原算法在激光器成本优化目标上的缺陷。随着准分子激光器及其剂量控制技术的进一步发展,新算法可望更能显示出其优越性并具有更广阔的应用前景。     

投影显示亮度均匀性稳健校正方法

随着投影仪的发展,其显示质量越来越受到人们关注.投影显示中普遍存在亮度不均匀性问题,且大部分高灰度图像在亮度校正后易出现饱和.而一般的剪切法会极大地降低像面均匀性.为解决这个问题,基于投影仪相机系统建立系统光度学模型,在传统线性映射方法的基础上提出了一种亮度不均匀性校正的非线性稳定算法.实验表明,采用的投影仪在校正前其图像显示亮度从中心到边缘下降了近50%,像面不均匀性的相对均方根误差(RRMS)为17.7%;而饱和图像校正后像面的RRMS为1.69%,几乎达到了与未饱和图像(1.55%)相同的校正精度,同时基本保持了显示图像的平均亮度水平.     

IMP重离子治癌中的剂量计算方法

针对深部肿瘤重离子治疗临床试验的需求,首先在水介质中进行生物有效剂量的优化计算,然后根据CT图像中像素CT值与水等效长度转换系数之间的关系, 结合水中的深度剂量分布曲线对每个像素进行积分得到CT图像上的生物有效剂量分布。同时介绍了基于被动式束流配送系统适形照射时的剂量确定方式, 并提出二维适形放疗也应使用分层照射方式以适应治疗时的不同要求。 这些方法适合目前及今后在IMP进行的重离子治癌临床试验研究中治疗计划系统的需要。Basic algorithms of biological effective dose optimization and dose distribution on CT image for the heavy ion therapy project at the Institute of Modern Physics (IMP), Chinese Academy of Sciences (CAS) are reported in this paper. Firstly, biological effective dose optimization is conducted in water. According to the relationship between CT number and water equivalent path length, an integral algorithm is used to calculate the average dose within a pixel and then the dose distribution in tissue is derived. Secondly, the dose determination of layer stacking conformal irradiation is described and the layer stacking method is proposed to be applied in two dimensional conformal irradiation. These methods are sufficient to the requirements of the ongoing and future heavy ion clinical trials conducted at IMP.     

一种稳健的道路主方向提取算法

基于高分辨遥感影像下直线型道路段的形状特征，提出了一种稳健的道路主方向提取算法．该算法基于结构张量计算的方向直方图探测初始主方向，并根据矢量化的边缘计算精确的方向角，使得探测过程稳健、结果准确．在此基础上可以发展交互式的直线道路段或道路网提取算法．实验结果表明，该算法的结果为进一步的道路边线的准确提取提供了保障．     

一种用于多叶准直器适形放疗的剂量算法研究

描述了一种用于多叶准直器适形放疗的剂量计算方法。 在不同大小的方野和非规则野照射情况下, 将此方法得到的剂量分布与微分卷积法计算得到的剂量分布进行了精度验证。 结果表明, 这种算法具有较高的剂量精度和较快的计算速度, 因此可在适形放疗中用作MLC适形野的自动生成。 A dose calculation algorithm for MLC based conformal radiotherapy is described in this paper. The algorithm is formulated by the coordinate of MLC leaves. Verification on the algorithm is made by comparing the dose distributions generated by this algorithm with that generated by a Differential Convolution Superposition algorithm for various regular and irregular fields. The results demonstrate that the present algorithm has suitable accuracy and high computational efficiency, thus it could be useful for the treatment planning process in MLC based conformal radiotherapy, where the workload for interactively or automatically designing the shapes of MLC is heavy.     

重离子治癌中的三维剂量成形方式

介绍了重离子治癌装置ＨＩＭＡＣ和ＨＩＴＡＧ的三维剂量成形方式;对一种正在设计中的重离子对肿瘤的照射方式进行了分析,旨在为ＨＩＲＦＬ及ＨＩＲＦＬ－ＣＳＲ所提供的离子束设计三维剂量成形方案． Three dimensional dose shaping methods in heavy-ion therapy facilities,HI MAC and HITAG are introduced.A new tumor irradiation pattern with heavy ion beam which is being designed is analyzed.It aims at choosing suitable three dimensional dose shaping planning for heavy ion beam supplied by HIRFL and the proposed HIRFL-CSR.     

弧形调强放射治疗对剂量计算方法的要求

在弧形调强放射治疗的治疗计划设计中, 由于包含有很多照射方向, 调强最优化的射束元矩阵计算需要很大的计算量和存储量, 为提高计算效率常使用简化剂量计算模型计算射束元矩阵, 因此有必要研究简化模型对治疗计划质量产生影响。 对一个模拟例子和一个临床实例, 使用没考虑散射效应的原射线剂量计算模型计算射束元矩阵, 由此进行最优化计算。 在得到最优化强度分布后, 通过比较原射线剂量计算模型和微分卷积剂量计算模型得到的剂量分布, 研究了不同射束数目条件下, 使用简化剂量计算模型计算射束元剂量矩阵对最终的剂量分布质量的影响。 结果表明, 在射线束很多的情况下(对应弧形调强照射）, 用简化的剂量计算模型, 即不考虑散射来计算射束元剂量矩阵, 会导致靶区剂量分布的质量大大低于预期的剂量分布质量, 因此, 弧形调强放射治疗的最优化计算中, 有效考虑散射的影响是必要的。 In the treatment planning for arc intensity modulated radiation therapy, because many irradiation directions are involved, the computing time and storage space needed for calculating beamlet dose matrices in optimization is quite heavy. In order to improve the computation efficiency, the simplified dose calculation is often used for the calculation of the dose matrices. Thus, it is deserved to study how this simplification could influence the quality of the treatment plan. In this paper, a simulation and a clinical case are adopted. Using the primary dose calculation model without taking into account the scattering effect to generate the dose matrices of beamlets, the optimization for beam intensity profile are firstly carried out. Then, based on the obtained intensity profile, the dose distributions are recalculated by using the primary dose calculation model and the differential convolution superposition dose calculation model which is more accurate but more time consuming. By comparing dose distributions obtained by this two models, the influence of using simplified model for dose matrix calculation on beam profile optimization is studied. The results demonstrate that when the beam number is large(corresponding to the arc modulated radiation), using the simplified model for the calculation of dose matrix of beamlets will reduce the quality of dose distribution greatly comparing with the expected dose distribution quality. Thus it is very necessary to correctly take into account the scattering effect in beam profile optimization for the arc intensity modulated radiation therapy.