Abstract：
OBJECTIVE: In contemporary radiotherapy, patient positioning accuracy relies on kV imaging. This study aims at optimizing planar kV image acquisition protocols regarding patient dose without degrading image quality.
METHODS: An image quality test-object was placed in-between PMMA plates, suitably arranged to model head or pelvis. Constructed phantoms were imaged using default protocols, the resultant image quality was assessed and the corresponding radiation dose was measured. The process was repeated using numerous kV/mAs combinations to identify those acquisition settings providing images at lower dose than the default protocols but without deterioration in image quality. Default and dose-optimized protocols were then tested on an anthropomorphic phantom and on 51 patients during two successive treatment sessions. Image quality was independently assessed by two readers. Organ and effective doses were estimated using a Monte Carlo simulation software.
RESULTS: Low-contrast detectability exhibited a stronger dependence on kV/mAs settings, compared to high-contrast resolution. Dose-optimized protocols resulted in significant dose reductions (anteroposterior-head 48.0 %, lateral-head 30.0 %, anteroposterior-pelvis 28.4 %, lateral-pelvis 27.0 %) compared to the default ones, without compromising image quality. Optimized protocols decreased effective doses by 54 % and 29.6 % in head and pelvic acquisitions, respectively. Regarding image quality, anthropomorphic and patient images acquired using the dose-optimized protocols were subjectively evaluated equivalent to those obtained with the corresponding default settings, indicating that the proposed protocols may be routinely used.
CONCLUSIONS: Given the potentially large number of radiotherapy fractions and the pertinent image acquisitions, dose-optimized protocols could significantly reduce patient dose associated with planar imaging without compromising positioning accuracy.
METHODS: An image quality test-object was placed in-between PMMA plates, suitably arranged to model head or pelvis. Constructed phantoms were imaged using default protocols, the resultant image quality was assessed and the corresponding radiation dose was measured. The process was repeated using numerous kV/mAs combinations to identify those acquisition settings providing images at lower dose than the default protocols but without deterioration in image quality. Default and dose-optimized protocols were then tested on an anthropomorphic phantom and on 51 patients during two successive treatment sessions. Image quality was independently assessed by two readers. Organ and effective doses were estimated using a Monte Carlo simulation software.
RESULTS: Low-contrast detectability exhibited a stronger dependence on kV/mAs settings, compared to high-contrast resolution. Dose-optimized protocols resulted in significant dose reductions (anteroposterior-head 48.0 %, lateral-head 30.0 %, anteroposterior-pelvis 28.4 %, lateral-pelvis 27.0 %) compared to the default ones, without compromising image quality. Optimized protocols decreased effective doses by 54 % and 29.6 % in head and pelvic acquisitions, respectively. Regarding image quality, anthropomorphic and patient images acquired using the dose-optimized protocols were subjectively evaluated equivalent to those obtained with the corresponding default settings, indicating that the proposed protocols may be routinely used.
CONCLUSIONS: Given the potentially large number of radiotherapy fractions and the pertinent image acquisitions, dose-optimized protocols could significantly reduce patient dose associated with planar imaging without compromising positioning accuracy.
摘要：
目标：在当代放射治疗中，患者定位精度依赖于kV成像。本研究旨在优化有关患者剂量的平面kV图像采集协议，而不会降低图像质量。
方法：将图像质量测试对象放置在PMMA板之间，适当地安排以模拟头部或骨盆。使用默认协议对构建的体模进行成像，评估所得图像质量并测量相应的辐射剂量。使用许多kV/mAs组合重复该过程,以识别在比默认方案更低的剂量下提供图像但图像质量没有劣化的那些采集设置。然后在两个连续的治疗过程中，对拟人化体模和51名患者进行了默认和剂量优化方案的测试。图像质量由两名读者独立评估。使用蒙特卡洛模拟软件估计器官和有效剂量。
结果：低对比度可检测性表现出对kV/mAs设置的更强依赖性，与高对比度分辨率相比。剂量优化方案导致剂量显著减少(前后头48.0%，侧头30.0%,前后骨盆28.4%，外侧骨盆27.0%)与默认骨盆相比，不影响图像质量。优化方案使头部和骨盆采集的有效剂量减少了54％和29.6％，分别。关于图像质量，使用剂量优化方案获得的拟人化和患者图像进行主观评估，等同于使用相应的默认设置获得的图像，表明拟议的协议可以常规使用。
结论：考虑到潜在的大量放疗部分和相关的图像采集，剂量优化方案可以显著减少与平面成像相关的患者剂量,而不会影响定位精度.
方法：将图像质量测试对象放置在PMMA板之间，适当地安排以模拟头部或骨盆。使用默认协议对构建的体模进行成像，评估所得图像质量并测量相应的辐射剂量。使用许多kV/mAs组合重复该过程,以识别在比默认方案更低的剂量下提供图像但图像质量没有劣化的那些采集设置。然后在两个连续的治疗过程中，对拟人化体模和51名患者进行了默认和剂量优化方案的测试。图像质量由两名读者独立评估。使用蒙特卡洛模拟软件估计器官和有效剂量。
结果：低对比度可检测性表现出对kV/mAs设置的更强依赖性，与高对比度分辨率相比。剂量优化方案导致剂量显著减少(前后头48.0%，侧头30.0%,前后骨盆28.4%，外侧骨盆27.0%)与默认骨盆相比，不影响图像质量。优化方案使头部和骨盆采集的有效剂量减少了54％和29.6％，分别。关于图像质量，使用剂量优化方案获得的拟人化和患者图像进行主观评估，等同于使用相应的默认设置获得的图像，表明拟议的协议可以常规使用。
结论：考虑到潜在的大量放疗部分和相关的图像采集，剂量优化方案可以显著减少与平面成像相关的患者剂量,而不会影响定位精度.
