联合单次散射模拟与蒙特卡罗模拟标定的散射校正改善了15O气体吸入脑部三维PET/CT成像的视觉伪影和定量准确性
该研究通过体模实验和脑血管病患者15O气体脑PET/CT成像,检验联合单次散射模拟与蒙特卡罗模拟计算标定因子的新散射校正方法(MCS-SSS)的有效性。体模实验中,MCS-SSS在各瓶放射性水平下误差在5%以内,而传统TFS-SSS在瓶放射性为2.4、5.1、9.7 kBq/mL时分别低估18%、36%、70%。患者成像中TFS-SSS图像出现冷伪影,而MCS-SSS图像未见伪影。
要点
- 体模实验中,瓶放射性为2.4、5.1、9.7 kBq/mL时,TFS-SSS图像活度浓度分别低估18%、36%、70%。
- MCS-SSS在各瓶放射性水平下误差在5%以内,但瓶放射性极高时图像在瓶周围出现轻度高活度伪影。
- 患者15O2和C15O2吸入成像中,TFS-SSS图像见冷伪影,MCS-SSS图像均未见伪影。
核心结论
- MCS-SSS在三维采集模式下能准确校正15O气体脑PET的散射,防止TFS-SSS图像中面罩旁出现的冷伪影。
- 作者认为MCS-SSS方法将有助于准确的定量评估。
研究设计与样本
- 研究包含体模实验和患者研究两部分。
- 体模实验中将模拟面罩的塑料瓶连接于模拟脑的圆柱体模,圆柱体模填充18F-FDG溶液(3.8-7.0 kBq/mL),瓶内填充非放射性空气或不同水平18F-FDG(0-170 kBq/mL)。
- 患者研究对脑血管病患者进行基于稳态法的15O气体脑PET,以获得脑血流量和氧代谢的定量图像。
主要发现
- 体模实验中,TFS-SSS图像在瓶旁出现冷伪影,瓶放射性为2.4、5.1、9.7 kBq/mL时,圆柱体模图像活度浓度分别低估18%、36%、70%。
- 瓶放射性更高时,圆柱体模图像活度浓度低估大于98%。
- MCS-SSS在各瓶放射性水平下误差在5%以内,但瓶含显著高放射性时图像在瓶周围出现轻度高活度伪影。
- 患者15O2和C15O2吸入成像中,TFS-SSS图像见冷伪影,MCS-SSS图像均未见伪影。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
In 3-dimensional PET/CT imaging of the brain with 15O-gas inhalation, high radioactivity in the face mask creates cold artifacts and affects the quantitative accuracy when scatter is corrected by conventional methods (e.g., single-scatter simulation [SSS] with tail-fitting scaling [TFS-SSS]). Here we examined the validity of a newly developed scatter-correction method that combines SSS with a scaling factor calculated by Monte Carlo simulation (MCS-SSS). Methods: We performed phantom experiments and patient studies. In the phantom experiments, a plastic bottle simulating a face mask was attached to a cylindric phantom simulating the brain. The cylindric phantom was filled with 18F-FDG solution (3.8-7.0 kBq/mL). The bottle was filled with nonradioactive air or various levels of 18F-FDG (0-170 kBq/mL). Images were corrected either by TFS-SSS or MCS-SSS using the CT data of the bottle filled with nonradioactive air. We compared the image activity concentration in the cylindric phantom with the true activity concentration. We also performed 15O-gas brain PET based on the steady-state method on patients with cerebrovascular disease to obtain quantitative images of cerebral blood flow and oxygen metabolism. Results: In the phantom experiments, a cold artifact was observed immediately next to the bottle on TFS-SSS images, where the image activity concentrations in the cylindric phantom were underestimated by 18%, 36%, and 70% at the bottle radioactivity levels of 2.4, 5.1, and 9.7 kBq/mL, respectively. At higher bottle radioactivity, the image activity concentrations in the cylindric phantom were greater than 98% underestimated. For the MCS-SSS, in contrast, the error was within 5% at each bottle radioactivity level, although the image generated slight high-activity artifacts around the bottle when the bottle contained significantly high radioactivity. In the patient imaging with 15O2 and C15O2 inhalation, cold artifacts were observed on TFS-SSS images, whereas no artifacts were observed on any of the MCS-SSS images. Conclusion: MCS-SSS accurately corrected the scatters in 15O-gas brain PET when the 3-dimensional acquisition mode was used, preventing the generation of cold artifacts, which were observed immediately next to a face mask on TFS-SSS images. The MCS-SSS method will contribute to accurate quantitative assessments.
引用
- Magota K, Shiga T, Asano Y, et al. Scatter Correction with Combined Single-Scatter Simulation and Monte Carlo Simulation Scaling Improved the Visual Artifacts and Quantification in 3-Dimensional Brain PET/CT Imaging with 15O-Gas Inhalation. J Nucl Med. 2017. PMID: 28646012. doi:10.2967/jnumed.117.193060
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。