Department of Biomedical Engineering, Duke University, Durham, NC 27705, USA. Carl E Ravin Advanced Imaging Laboratories, Duke University, Durham, NC 27705, USA.
Phys Med Biol. 2013 Dec 21;58(24):8755-68. doi: 10.1088/0031-9155/58/24/8755. Epub 2013 Dec 4.
Recent studies have shown the feasibility of estimating patient dose from a CT exam using CTDI(vol)-normalized-organ dose (denoted as h), DLP-normalized-effective dose (denoted as k), and DLP-normalized-risk index (denoted as q). However, previous studies were limited to a small number of phantom models. The purpose of this work was to provide dose coefficients (h, k, and q) across a large number of computational models covering a broad range of patient anatomy, age, size percentile, and gender. The study consisted of 100 patient computer models (age range, 0 to 78 y.o.; weight range, 2-180 kg) including 42 pediatric models (age range, 0 to 16 y.o.; weight range, 2-80 kg) and 58 adult models (age range, 18 to 78 y.o.; weight range, 57-180 kg). Multi-detector array CT scanners from two commercial manufacturers (LightSpeed VCT, GE Healthcare; SOMATOM Definition Flash, Siemens Healthcare) were included. A previously-validated Monte Carlo program was used to simulate organ dose for each patient model and each scanner, from which h, k, and q were derived. The relationships between h, k, and q and patient characteristics (size, age, and gender) were ascertained. The differences in conversion coefficients across the scanners were further characterized. CTDI(vol)-normalized-organ dose (h) showed an exponential decrease with increasing patient size. For organs within the image coverage, the average differences of h across scanners were less than 15%. That value increased to 29% for organs on the periphery or outside the image coverage, and to 8% for distributed organs, respectively. The DLP-normalized-effective dose (k) decreased exponentially with increasing patient size. For a given gender, the DLP-normalized-risk index (q) showed an exponential decrease with both increasing patient size and patient age. The average differences in k and q across scanners were 8% and 10%, respectively. This study demonstrated that the knowledge of patient information and CTDIvol/DLP values may be used to estimate organ dose, effective dose, and risk index in abdominopelvic CT based on the coefficients derived from a large population of pediatric and adult patients.
最近的研究表明，使用 CTDI(vol)-标准化器官剂量（表示为 h）、DLP-标准化有效剂量（表示为 k）和 DLP-标准化风险指数（表示为 q），从 CT 检查中估算患者剂量是可行的。然而，以前的研究仅限于少数体模模型。本工作的目的是提供涵盖广泛患者解剖、年龄、大小百分位和性别范围的大量计算模型的剂量系数（h、k 和 q）。该研究包括 100 例患者计算机模型（年龄范围 0 至 78 岁；体重范围 2-180 千克），其中包括 42 例儿科模型（年龄范围 0 至 16 岁；体重范围 2-80 千克）和 58 例成人模型（年龄范围 18 至 78 岁；体重范围 57-180 千克）。研究纳入了来自两家商业制造商的多探测器阵列 CT 扫描仪（GE Healthcare 的 LightSpeed VCT 和 Siemens Healthcare 的 SOMATOM Definition Flash）。使用经过验证的蒙特卡罗程序模拟每个患者模型和每个扫描仪的器官剂量，从中得出 h、k 和 q。确定了 h、k 和 q 与患者特征（大小、年龄和性别）之间的关系。进一步描述了扫描仪之间转换系数的差异。CTDI(vol)-标准化器官剂量（h）随患者体型增大呈指数下降。对于图像覆盖范围内的器官，跨扫描仪的 h 平均差异小于 15%。对于图像覆盖范围之外或周边的器官，该值增加到 29%，对于分布器官，该值增加到 8%。DLP-标准化有效剂量（k）随患者体型增大呈指数下降。对于给定的性别，DLP-标准化风险指数（q）随患者体型和年龄增大呈指数下降。跨扫描仪的 k 和 q 的平均差异分别为 8%和 10%。本研究表明，基于来自大量儿科和成年患者的系数，了解患者信息和 CTDIvol/DLP 值可能用于估算腹部和盆腔 CT 的器官剂量、有效剂量和风险指数。
