Mutanga Theodore F, de Boer Hans C J, van der Wielen Gerard J, Wentzler Davy, Barnhoorn Jaco, Incrocci Luca, Heijmen Ben J M
Department of Radiation Oncology, Division of Medical Physics, Erasmus Medical Center-Daniel den Hoed Oncology Center, Rotterdam, The Netherlands.
Int J Radiat Oncol Biol Phys. 2008 Jul 15;71(4):1074-83. doi: 10.1016/j.ijrobp.2007.11.022. Epub 2008 Jan 22.
A fully automated, fast, on-line prostate repositioning scheme using implanted markers, kilovoltage/megavoltage imaging, and remote couch movements has been developed and clinically applied. The initial clinical results of this stereographic targeting (SGT) method, as well as phantom evaluations, are presented.
Using the SGT method, portal megavoltage images are acquired with the first two to six monitor units of a treatment beam, immediately followed by acquisition of an orthogonal kilovoltage image without gantry motion. The image pair is automatically analyzed to obtain the marker positions and three-dimensional prostate displacement and rotation. Remote control couch shifts are applied to correct for the displacement. The SGT performance was measured using both phantom images and images from 10 prostate cancer patients treated using SGT.
With phantom measurements, the accuracy of SGT was 0.5, 0.2, and 0.3 mm (standard deviation [SD]) for the left-right, craniocaudal, and anteroposterior directions, respectively, for translations and 0.5 degrees (SD) for the rotations around all axes. Clinically, the success rate for automatic marker detection was 99.5%, and the accuracy was 0.3, 0.5 and 0.8 mm (SD) in the left-right, craniocaudal, and anteroposterior axes. The SDs of the systematic center-of-mass positioning errors (Sigma) were reduced from 4.0 mm to <0.5 mm for all axes. The corresponding SD of the random (sigma) errors was reduced from 3.0 to <0.8 mm. These small residual errors were achieved with a treatment time extension of <1 min.
Stereographic targeting yields systematic and random prostate positioning errors of <1 mm with <1 min of added treatment time.
已开发出一种使用植入标记物、千伏/兆伏成像和远程治疗床移动的全自动、快速在线前列腺重新定位方案，并已临床应用。本文介绍了这种立体定向靶向（SGT）方法的初步临床结果以及模体评估。
使用SGT方法，在治疗束的前两到六个监测单位采集门静脉兆伏图像，随后立即在机架不动的情况下采集正交千伏图像。自动分析该图像对以获得标记物位置以及前列腺的三维位移和旋转。应用远程控制治疗床移位来校正位移。使用模体图像和10例接受SGT治疗的前列腺癌患者的图像来测量SGT性能。
通过模体测量，SGT在左右、头脚和前后方向上的平移精度分别为0.5、0.2和0.3毫米（标准差[SD]），在所有轴向上的旋转精度为0.5度（SD）。临床上，自动标记物检测的成功率为99.5%，在左右、头脚和前后轴上的精度分别为0.3、0.5和0.8毫米（SD）。所有轴向上系统质心定位误差（Sigma）的标准差从4.0毫米降至<0.5毫米。随机（sigma）误差的相应标准差从3.0降至<0.8毫米。在治疗时间延长<1分钟的情况下实现了这些小的残余误差。
立体定向靶向在增加<1分钟治疗时间的情况下，产生的系统和随机前列腺定位误差<1毫米。
