Dehairs M, Bosmans H, Desmet W, Marshall N W
Medical Imaging Research Centre, Medical Physics and Quality Assessment, Katholieke Universiteit Leuven, 3000 Leuven, Belgium.
Phys Med Biol. 2017 Jul 31;62(16):6610-6630. doi: 10.1088/1361-6560/aa7a9d.
Current automatic dose rate controls (ADRCs) of dynamic x-ray imaging systems adjust their acquisition parameters in response to changes in patient thickness in order to achieve a constant signal level in the image receptor. This work compares a 3 parameter (3P) ADRC control to a more flexible 5-parameter (5P) method to meet this goal. A phantom composed of 15 composite poly(methyl) methacrylate (PMMA)/aluminium (Al) plates was imaged on a Siemens Artis Q dynamic system using standard 3P and 5P ADRC techniques. Phantom thickness covered a water equivalent thickness (WET) range of 2.5 cm to 37.5 cm. Acquisition parameter settings (tube potential, tube current, pulse length, copper filtration and focus size) and phantom entrance air kerma rate (EAKR) were recorded as the thickness changed. Signal difference to noise ratio (SDNR) was measured using a 0.3 mm iron insert centred in the PMMA stack, positioned at the system isocentre. SDNR was then multiplied by modulation transfer function (MTF) based correction factors for focal spot penumbral blurring and motion blurring, to give a spatial frequency dependent parameter, SDNR(u). These MTF correction factors were evaluated for an object motion of 25 mm s and at a spatial frequency of 1.4 mm in the object plane, typical for cardiac imaging. The figure of merit (FOM) was calculated as SDNR(u)²/EAKR for the two ADRC regimes. Using 5P versus 3P technique showed clear improvements over all thicknesses. Averaged over clinically relevant adult WET values (20 cm-37.5 cm), EAKR was reduced by 13% and 27% for fluoroscopy and acquisition modes, respectively, while the SDNR(u) based FOM increased by 16% and 34% for fluoroscopy and acquisition. In conclusion, the generalized FOM, taking into account the influence of focus size and object motion, showed benefit in terms of image quality and patient dose for the 5-parameter control over 3-parameter method for the ADRC programming of dynamic x-ray imaging systems.
当前动态X射线成像系统的自动剂量率控制（ADRC）会根据患者厚度的变化调整采集参数，以便在图像接收器中实现恒定的信号水平。这项工作将一种三参数（3P）ADRC控制与一种更灵活的五参数（5P）方法进行比较，以实现这一目标。使用标准的3P和5P ADRC技术，在西门子Artis Q动态系统上对一个由15块复合聚甲基丙烯酸甲酯（PMMA）/铝（Al）板组成的体模进行成像。体模厚度覆盖了2.5厘米至37.5厘米的水等效厚度（WET）范围。随着厚度变化，记录采集参数设置（管电压、管电流、脉冲长度、铜过滤和焦点尺寸）以及体模入口空气比释动能率（EAKR）。使用一个位于PMMA堆栈中心、置于系统等中心的0.3毫米铁插入物测量信号差噪比（SDNR）。然后，将SDNR乘以基于调制传递函数（MTF）的焦点半影模糊和运动模糊校正因子，得到一个空间频率相关参数SDNR(u)。这些MTF校正因子是针对25毫米/秒的物体运动以及物体平面中1.4毫米的空间频率进行评估的，这对于心脏成像来说是典型的。针对两种ADRC方案，品质因数（FOM）计算为SDNR(u)²/EAKR。与3P技术相比，5P技术在所有厚度上都有明显改进。在临床相关的成人WET值（20厘米至37.5厘米）范围内进行平均，荧光透视和采集模式下的EAKR分别降低了13%和27%，而基于SDNR(u)的FOM在荧光透视和采集中分别提高了16%和34%。总之，考虑到焦点尺寸和物体运动的影响，广义FOM表明，对于动态X射线成像系统的ADRC编程，五参数控制在图像质量和患者剂量方面优于三参数方法。
