Department of Radiation Oncology, Columbia University Irving Medical Center, New York, NY, USA.
Department of Neurological Surgery, Columbia University Irving Medical Center, New York, NY, USA.
Med Phys. 2019 Oct;46(10):4333-4339. doi: 10.1002/mp.13740. Epub 2019 Aug 20.
Cone beam computed tomography (CBCT) imaging has been implemented on the Leksell Gamma Knife Icon™ for assessing patient positioning in mask-based Gamma Knife radiosurgery. The purpose of this study was to evaluate the performance of the CBCT-based patient positioning system as a tool for frameless Gamma Knife radiosurgery.
Daily quality assurance (QA) CBCT precision test results from a 12-month period were analyzed for the geometric accuracy and the stability of the imager. The performance of the image acquisition module and the image registration algorithm was evaluated using an anthropomorphic head phantom (CIRS Inc., Norfolk, VA) and a XYZR axis manual positioning stage (TOAUTO Inc., Guangdong, China). The head phantom was fixed on a mask adaptor and manually translated in the X, Y, Z directions or rotated around the X, Y, Z axes in the range of ±10 mm or ±10º. A CBCT scan was performed after each manual position setup followed by an image registration to the reference scan. To assess the overall setup uncertainties in fractionated treatment, two cylindrical Presage phantoms (Heuris Inc., Skillman, NJ) of 15 cm diameter and 10 cm height were irradiated with identical prescription dose and shot placement following standard mask-based treatment workflow according to two different fraction schedules: a single fraction treatment of 7.5 Gy and a 5-fraction treatment with 1.5 Gy per fraction.
The averaged vector deviations of the four marks from their preset values are 0.087, 0.085, 0.095, and 0.079 mm from the 212 daily QA tests. The averaged displacements in the X, Y, Z coordinates and the pitch, yaw, roll angles from the image registration tests are 0.23, 0.27, 0.14, 0.32º, 0.19º, 0.31º from the manual setup. The corresponding maximum differences are 0.41, 0.33, 0.29 mm, 0.45º, 0.31º, and 0.43º, respectively. Compared to the treatment plan using the 2% & 1 mm criteria, the averaged 2D Gamma passing rate is 98.25% for the measured dose distribution from the Presage phantom with 1-fraction irradiation and 95.12% for the 5-fraction irradiation. The averaged Gamma passing rates are 99.53% and 98.16% for the 1-fraction and 5-fraction irradiations using the 2% & 2 mm criteria.
The CBCT imager and the image registration algorithm can reproduce phantom position with <0.5 mm/0.5º uncertainty. A systematic contribution from the interfraction phantom repositioning procedure was observed in the Gamma analysis over the irradiated volumes of two end-to-end test phantoms.
锥形束 CT（CBCT）成像已应用于 Leksell Gamma Knife Icon™，用于评估基于面罩的伽玛刀放射外科中的患者定位。本研究的目的是评估基于 CBCT 的患者定位系统作为无框架伽玛刀放射外科工具的性能。
分析了 12 个月的日常质量保证（QA）CBCT 精度测试结果，以评估成像仪的几何精度和稳定性。使用人体头部模拟体模（CIRS Inc.，弗吉尼亚州诺福克）和 XYZR 轴手动定位台（TOAUTO Inc.，中国广东）评估图像采集模块和图像配准算法的性能。头部模拟体模固定在面罩适配器上，并手动沿 X、Y、Z 方向平移或绕 X、Y、Z 轴旋转±10mm 或±10°。每次手动定位设置后进行一次 CBCT 扫描，然后与参考扫描进行图像配准。为了评估分次治疗中的整体设置不确定性，根据两种不同的分次方案，对两个直径为 15cm、高度为 10cm 的圆柱形 Presage 模拟体（Heuris Inc.，新泽西州 Skillman）进行了相同的处方剂量和射野放置照射：单次 7.5Gy 治疗和 5 次 1.5Gy 治疗，每次治疗。
从 212 次日常 QA 测试中，四个标记相对于其预设值的平均向量偏差为 0.087、0.085、0.095 和 0.079mm。从图像配准测试中，X、Y、Z 坐标和俯仰、偏航、滚动角度的平均位移为 0.23、0.27、0.14、0.32°、0.19°、0.31°，手动设置。相应的最大差异分别为 0.41、0.33、0.29mm、0.45°、0.31°和 0.43°。与使用 2%和 1mm 标准的治疗计划相比，使用 1 次照射的 Presage 模拟体测量的剂量分布的平均 2D Gamma 通过率为 98.25%，5 次照射的平均 Gamma 通过率为 95.12%。使用 2%和 2mm 标准，单次和 5 次照射的平均 Gamma 通过率分别为 99.53%和 98.16%。
CBCT 成像仪和图像配准算法可以以<0.5mm/0.5°的不确定性重现体模位置。在两个端到端测试体模的照射体积上的伽马分析中观察到分次体模重新定位过程的系统贡献。
