动态楔形因子的研究

目的:在1978年,P.K.Kijewski等人[1]提出动态楔形技术(DW)之后,20世纪90年代,这项技术开始应用于Varian加速器上,并根据Varian加速器的特性给出了求增强型动态楔形因子(EDWF)的公式[2].然而,这个公式是否能应用于所有的医用直线加速器呢?方法:以Siemens Primus医用直线加速器为对象进行了实验验证.对于其他类型的加速器,如果公式适用,公式中所出现的五个待定参量α0,α1,b1,α,β是否需要重新修正呢?结果:通过实验发现,Varian加速器的动态楔形因子的计算公式及公式中出现的参量用于Siemens Primus医用直线加速器时,将会出现3%误差.结论:用通过实验按照Siemens Primus加速器特性重新拟出的修正公式和修正参数则可以把误差控制在1%范围内.     

楔形野剂量计算中的误差分析和修正

目的研究楔形野剂量计算中的误差,并探讨解决方法.材料与方法在10MV和6MVX线条件下,用NEFarmer25710.6cc指形电离室和三维水箱在水模中测出平野和楔形野的各种参数,并用二种方法计算剂量,结果与实侧值比较.结果实测数据显示Pdd和Scp在平野和楔形野情况下存在差异.楔形因子因此随深度而变化,变化程度受射线能量、楔形板规格影响.与实测值比较,用传统方法计算楔形野剂量的结果存在误差,误差大小与能量、野面积、深度有关.6MVX线、15×15野、20cm深度处的计算误差可达11%.而用改进的方法进行计算,可将误差控制在1%以内.结论由于忽略了Pdd等物理参数在楔形野条件下的变化,用传统方法计算楔形野剂量存在误差.为保证临床剂量计算的准确性,应在计算公式中加入修正因子.     

6MV-15MV下物理楔形因子和动态楔形因子的比较

目的:探讨不同能量下,Varian21EX直线加速器中物理楔形因子和动态楔形因子受照射野大小和深度的影响。方法:在固体水膜体中利用0.6 cc电离室对6 MV和15 MV射线束下不同角度物理楔形板和动态楔形板分别测量加和不加楔形滤片时的剂量率来计算楔形因子。通过测量不同角度的物理楔形板和动态楔形板在固定照射野(10 cm×10 cm)的不同深度下的楔形因子来研究楔形因子随深度的变化规律。同时,对于楔形因子随射野大小的变化规律,还测量了不同角度的物理楔形板和动态楔形板在固定深度(d=10 cm)下的不同射野大小的楔形因子。为了更好地分析物理楔形因子与动态楔形因子的差异,引入了相对楔形因子NWF。结果:深度对于物理楔形板的楔形因子较为明显,深度增加时楔形因子增大,且随着楔形角的增大变化更明显。对于150、300、450、600的物理楔形板,当深度由最大深度增加到20 cm时对于6 MV能量楔形因子分别增加了1.86%、3.79%、4.99%、7.95%;对于15 MV能量1.29%、1.35%、1.49%、2.03%。而动态楔形因子随深度变化不明显,最大变化不到1%。射野大小对于物理楔形因子也有一定的影响,楔形因子随射野增加而增加,但是增加幅度不大;而对于动态楔形板,在6 MV和15 MV射线束下楔形因子受射野的增大都有明显的减小。对于100、150、200、250、300、450、600的动态楔形板,从参考射野(10 cm×10 cm)到最大射野,楔形因子分别减少了7.91%、11.04%、14.08%、16.96%、19.7%、28.03%、35.89%对于6 MV和5.72%、8.17%、10.41%、12.85%、15.08%、21.82%、30.59%对于15 MV能量。结论:对于物理楔形板,深度和射野大小都对物理楔形因子有影响,所以临床剂量计算时必须考虑深度和射野大小对物理楔形因子的影响并对它进行修正。对于动态楔形板,深度对动态楔形因子影响较小,在临床剂量计算时可以忽略;而射野大小对动态楔形因子影响比较明显,在临床剂量计算时只须考虑相对射野楔形因子。     

直线加速器楔形因子的影响因素和修正方法

目的 研究直线加速器楔形因子的影响因素,并提出可行的修正方法,为临床准确使用该因子提供依据.方法 对SIEMENS MD7745直线加速器和GE Saturne41直线加速器,6MV-X线,60°楔形滤片,测量不同大小的照射野且不同深度处加和不加楔形滤片时的剂量率,计算楔形因子.然后分析数据提出假设公式.结果 照射野大小和测量深度对楔形因子Fw均有影响:其中照射野的影响不大,可通过对大中小野取平均值的方法将精度控制在1%以内;但深度的影响却很大,必须通过修正公式Fw(d)=(a bd)进行修正才能达到WHO放射治疗质量保证和质量控制有关楔形因子的精度必须好于2%的规定.     

动态楔形板与物理楔形板剂量学的比较研究

目的:比较动态楔形板与物理楔形板剂量曲线分布的特点;方法:用二维电离室矩阵分别测量动态楔形板Y1-IN方向15°,30°,45°,60°;固定楔形板IN方向15°,30°,45°,60°,得到相应楔形野的注量图;Dose1剂量仪测量相同楔形角的动态楔形板与物理楔形板5cm深度的绝对剂量值;结果:相同楔形角的动态楔形板和物理楔形板的楔形曲线大致重合,同深度相同角度的动态楔形板比物理楔形板的绝对剂量值大15%～25%;结论:动态楔形板完全可以替代物理楔形板,提高工作效率和机器使用率。     

医用直线加速器输出剂量稳定性分析

目的：分析医用直线加速器输出剂量稳定性及其影响因素。方法：采用SPSS15.0统计分析软件,统计2009年每日治疗病人前监测6 MV、15 MVX射线,和9 MeV、12 MeV电子线输出剂量数据,分析医用直线加速器不同能量输出剂量的稳定性及其影响因素,提出加速器输出剂量质量保证的相关措施。结果：4档能量中的3档能量（9 MeV,12 MeV,15MV）输出剂量K-S检测双尾渐进概率P值分别为0.428、0.933、0.355均大于显著性水平0.05,符合正态分布。由于加速器微波源输出不稳定,6 MV输出剂量1月到3月,从98.4%连续不断漂移上升到102.5%。6 MV K-S检测双尾渐进概率P值是0.012小于显著性水平0.05,不符合正态分布。结论：直线加速器输出剂量的稳定性是肿瘤放射治疗治疗质量保证的重要方面。每日治疗肿瘤病人前监测和直线加速器输出剂量,分析直线加速器输出剂量的稳定性,有助于降低加速器系统误差,提高患者治疗剂量的精度。     

动态楔形板和物理楔形板对射线深度剂量与射野外周边剂量的影响

目的：探讨动态楔形板和物理楔形板对射线深度剂与射野外周边剂量的影响。方法：利用电离室法测量平野、动态楔形野、物理楔形野的深度剂量和射野外周边剂量。结果：动态楔形野的深度剂量和射野外周边剂量接近于平野的深度剂量和射野外周边剂量；而物理楔形野的深度剂量和射野外周边剂量高于平野的深度剂量和射野外周边剂量。结论：剂量计算时、动态楔形野可以利用平野的深度剂量，而物理楔形野需采用楔形深度剂量；使用物理楔形板时应注意想邻野或野外敏感器官的受量。     

Elekta Motorized Wedge 6MV X-ray楔形因子特性的初步研究

目的：探讨Elekta motorized wedge楔形因子随射野宽度和测量深度的变化特性。方法：对Elekta Precise直线加速器6 MV X-ray,用Farmer 2571指形电离室和美国Capintec 192剂量仪,在固定测量深度的条件下,逐步扩大射野,实测获得15°,30°,45°,60°四个角度楔形板的楔形因子随射野宽度的变化特性;在固定射野宽度的条件下,逐步改变测量点的深度,实测获得15°,30°,45°,60°四个角度楔形板的楔形因子随测量深度的变化特性;同时,将每个实测到得的楔形因子与Elekta Precise TPS 2.12模拟实测条件输出的楔形因子进行了对比。结果：Elekta motorized wedge楔形因子随射野宽度和测量深度的增加而变大,呈现线性变化。当FSZ〈20 cm×20 cm时,楔形因子随射野宽度线性变化的斜率比较大,当FSZ〉20 cm×20 cm时,楔形因子随射野宽度线性变化的斜率比较小,深度对楔形因子的影响小于射野宽度。Elekta Pre-cise TPS 2.12模拟实测条件输出的楔形因子与实测得到的相近,偏差较小。结论：当FSZ〈20 cm×20 cm时,宽度对楔形因子的影响不能忽略,因此处方剂量计算时应先求得等效方野,而后用该等效方野对应的楔形因子进行楔形野的处方剂量计算;当FSZ〉20 cm×20 cm时,可以采用20 cm×20 cm测得的楔形因子进行楔形野的处方剂量计算;深度对楔形因子的影响可忽略,可以将参考深度（水下10 cm）获得的楔形因子用于所有的深度。     

光子射野剂量计算参数研究

目的：介绍医用加速器常规光子射线的机器数据测量方法及剂量计算模型中基本参数的计算过程。以百分深度剂量与散射因子为基础数据,根据原散射线模型通过测量数据推导出原射线组织最大剂量比、散射最大剂量比、原射线在水中线性衰减系数、能量注量等,为进一步还原射野在水模体中的剂量分布提供方法与理论。方法：用Blue Phantom三维水箱在医科达Synergy加速器上测量6MV光子线的百分深度剂量、离轴比剂量、总散射因子、准直器散射因子,先从测量的百分深度剂量曲线中按照原散射模型剥离出原射线百分深度剂量,然后在Matlab软件中拟合处理测量的散射因子数据,外推出零野的模体散射因子,从而按照给定公式计算出组织最大剂量比、散射最大剂量比。按照离轴比剂量,利用平方反比规律推出最大开野在模体表面的能量注量。结果：计算出准直器散射因子、总散射因子的拟合公式,外推零野模体散射因子（s。）、根据原射线的百分深度剂量曲线计算出原射线在水中线性衰减系数,组织最大剂量比（TMR）、散射最大剂量比（SMR）、以及射野能量注量分布（Fluence Matrix）。结论：这些基本参数是剂量计算建模的关键,也是进一步研究各种剂量计算模型的基础。     

利用百分深度剂量计算组织最大剂量比的准确性研究

目的：研究用测量的X线照射野百分深度剂量和体模散射输出因子计算组织最大剂量比的可行性。方法：用PTwmp3三维水箱分别测量Precise加速器的6MV和10MVX线的百分深度剂量、组织最大剂量比以及照射野输出因子。利用NE2570剂量仪和自制的圆柱形有机玻璃体模测量加速器准直系统散射输出因子。用VisualBasic6．0编程计算组织最大剂量比，并将组织最大剂量比的计算值和测量值进行比较。结果：组织最大剂量比和射线能量、照射野面积有关。6MV和10MVX线的组织最大剂量比的计算值和测量值的误差小于2％。结论：组织最大剂量比的计算值和测量值符合得很好，可以直接应用于吸收剂量计算。     

Study and evaluation of the Siemens virtual wedge factor: dosimetric monitor system and variable field effects

In the year 1997 Siemens introduced the virtual wedge in its accelerators. The idea was that a dose profile similar to that of a physical wedge can be obtained by moving one of the accelerator jaws at a constant speed while the dose rate is changing. This work explores the observed behaviour of virtual wedge factors. A model is suggested which takes into account that at any point in time, when the jaw moves, the dose at a point of interest in the phantom is not only due to the direct beam. It also depends on the scattered radiation in the phantom, the head scatter and the behaviour of the monitoring system of the accelerator. Measurements are performed in a Siemens Primus accelerator and compared to the model predictions. It is shown that the model agrees reasonably well with measurements spanning a wide range of conditions. A strong dependence of virtual wedge factors on the dosimetric board has been confirmed and an explanation has been given on how the balance between different contributions is responsible for virtual wedge factors values.     

Suppression of dark current radiation in step-and-shoot intensity modulated radiation therapy by the initial pulse-forming network

The effect of the initial pulse forming network (IPFN) on the suppression of dark current is investigated for a Siemens Primus accelerator. The dark current produces a spurious radiation, which is referred to as dark current radiation (DCR) in this study. In the step-and-shoot delivery of an intensity modulated radiation therapy (IMRT), the DCR could be of some concern for whole body dose along with leakage radiation through collimator jaws or multileaf collimator. By adjusting the IPFN-to-PFN ratio to >0.8, the DCR can be measured with an ion chamber during the "PAUSE" state of the accelerator in the IMRT mode. For 15 MV x rays, the magnitude of the DCR is approximately equal to 0.7% of the dose at dmax for a 10 x 10 cm2 field. The DCR has a similar central axis depth dose as a 15 MV beam as determined from a water phantom scan. When the IPFN-to-PFN ratio is lowered to <0.8, no DCR is detected. For low energy x rays (6 MV), no DCR is detected regardless of the IPFN-to-PFN ratio. Although the DCR is studied only for the Siemens Primus model accelerator, the same precaution applies to other models of modern accelerators from other vendors. Due to the large number of field segments used in a step-and-shoot IMRT, it is imperative therefore, that dark current evaluation be part of machine commissioning and annual calibration for high-energy photon beams. Should DCR be detected, the medical physicist should work with a service engineer to rectify the problem. In view of DCR and whole body dose, low-energy photon beams are advisable for IMRT.     

Extension of CadPlan algorithm to model the dose distribution under a motorized wedge.

The CadPlan treatment planning system models the dose distribution in the non-wedge direction under a wedged field by converting the wedge thickness to an equivalent water thickness. The algorithm estimates the off-axis ratio (OAR) in the non-wedged direction using the open field OAR at a depth deeper by this equivalent water thickness. This model has been shown to work well for a Siemens Mevatron KD-2 Linac. However, the motorized wedge of the Elekta (formerly Philips) accelerators is tapered off-axis to give a flat dose profile in the non-wedged direction. The CadPlan model assumes that the wedge has a uniform thickness in the non-wedged direction and so cannot model the off-axis dose for the motorized wedge. For a 4 MV beam of a SL75/5 accelerator this leads to a 7% overestimate and a 9% underestimate of the OAR under the thin and thick edge of the wedge respectively. For 6 and 18 MV beams of a SL20 accelerator and a 6 MV beam of a SL75/5 accelerator, the model underestimates the OAR in the order of 10% under the thick end of the wedge. We have shown that by appropriate modification of the effective water thickness values at off-axis distances, the algorithm models the OAR in the non-wedged direction to within 2.5% of the measured values for the 4, 6 and 18 MV beams, for the Elekta motorized wedge.     

Photon spectral characteristics of dissimilar 6 MV linear accelerators

This work measures and compares the energy spectra of four dosimetrically matched 6 MV beams, generated from four physically different linear accelerators. The goal of this work is twofold. First, this study determines whether the spectra of dosimetrically matched beams are measurably different. This study also demonstrates that the spectra of clinical photon beams can be measured as a part of the beam data collection process for input to a three-dimensional (3D) treatment planning system. The spectra of 6 MV beams that are dosimetrically matched for clinical use were studied to determine if the beam spectra are similarly matched. Each of the four accelerators examined had a standing waveguide, but with different physical designs. The four accelerators were two Varian 2100C/Ds (one 6 MV/18 MV waveguide and one 6 MV/10 MV waveguide), one Varian 600 C with a vertically mounted waveguide and no bending magnet, and one Siemens MD 6740 with a 6 MV/10 MV waveguide. All four accelerators had percent depth dose curves for the 6 MV beam that were matched within 1.3%. Beam spectra were determined from narrow beam transmission measurements through successive thicknesses of pure aluminum along the central axis of the accelerator, made with a graphite Farmer ion chamber with a Lucite buildup cap. An iterative nonlinear fit using a Marquardt algorithm was used to find each spectrum. Reconstructed spectra show that all four beams have similar energy distributions with only subtle differences, despite the differences in accelerator design. The measured spectra of different 6 MV beams are similar regardless of accelerator design. The measured spectra show excellent agreement with those found by the auto-modeling algorithm in a commercial 3D treatment planning system that uses a convolution dose calculation algorithm. Thus, beam spectra can be acquired in a clinical setting at the time of commissioning as a part of the routine beam data collection.     

Comparison of dosimetric characteristics of Siemens virtual and physical wedges

Dosimetric properties of Virtual Wedge (VW) and physical wedge (PW) in 6 and 23 MV photon beams from a Siemens Primus linear accelerator, including wedge factors, depth doses, dose profiles, peripheral doses and surface doses, are compared. While there is a great difference in absolute values of wedge factors, VW factors (VWFs) and PW factors (PWFs) have a similar trend as a function of field size. PWFs have a stronger depth dependence than VWF due to beam hardening in PW fields. VW dose profiles in the wedge direction, in general, match very well with PW, except in the toe area of large wedge angles with large field sizes. Dose profiles in the nonwedge direction show a significant reduction in PW fields due to off-axis beam softening and oblique filtration. PW fields have significantly higher peripheral doses than open and VW fields. VW fields have similar surface doses as the open fields while PW fields have lower surface doses. Surface doses for both VW and PW increase with field size and slightly with wedge angle. For VW fields with wedge angles 45 degrees and less, the initial gap up to 3 cm is dosimetrically acceptable when compared to dose profiles of PW. VW fields in general use less monitor units than PW fields.     

Dosimetric aspects of a 3.3-MV linear accelerator

Both the design considerations and the dosimetric properties of the Siemens Model 5800 linear accelerator are discussed. This unit is of such an energy (3.3 MV) as to imitate Cobalt-60 teletherapy depth doses. A linear relation of dmax to depth dose at low energies was found for various wave guides and targets. The energy of the unit can be characterized by its nominal accelerating potential of 2.70 MV, its d80 of 5.3 cm, its first half-value layer of 0.8 cm lead and the measured energy of the electron beam at 3.3 MeV. The following selected commissioning aspects are reported: central axis depth dose, relative output factors, beam profiles, wedge factors, virtual source position, back scatter factors, penumbra and build-up region.     

Monte Carlo modelling of a virtual wedge

Compared with a set of physical photon wedges, a non physical wedge (virtual or dynamic wedge), realized by a moving collimator jaw, offers an alternative that allows creation of a wedged field with any arbitrary wedge angle instead of the traditional four physical wedges (15 degrees, 30 degrees, 45 degrees and 60 degrees). It is commonly assumed that non-physical wedges do not alter the photon spectrum compared with physical wedges that introduce beam hardening and loss of dose uniformity in the unwedged direction. In this study, we investigated the influence of a virtual wedge on the photon spectra of a 6-10 MV Siemens MD2 accelerator with the Monte Carlo code EGS4/BEAM. Good agreement was obtained between calculated and measured lateral dose profiles at the depth of maximum dose and at 10 cm depth for 20 x 20 cm2 fields for 6 and 10 MV photon beams. By comparing Monte Carlo models of a physical wedge and the virtual wedge that was studied in this work, it is confirmed that the latter has an insignificant effect on the beam quality, whereas the former can introduce significant beam hardening.