Comparison of Different Attenuation Correction Methods for Dual Gating Myocardial Perfusion SPECT/CT

Comparison of Different Attenuation Correction Methods for Dual Gating Myocardial Perfusion SPECT/CT
复制标题

双门控心肌灌注 SPECT/CT 不同衰减校正方法的比较

DOI:
10.1109/trpms.2019.2899066
复制
发表时间:
2019-09-01
影响因子:
4.4
通讯作者:
Mok, Greta S. P.
Mok, Greta S. P.
中科院分区:
其他
文献类型:
--
作者:
Zhang, Qi;Zhang, Duo;Mok, Greta S. P.

文献摘要

被引文献

相似文献

双呼吸-心脏门控实现运动冻结心肌灌注单光子发射计算机断层扫描(SPECT)。本文的目的是评估双门控心脏 SPECT 的各种衰减图的性能。我们使用具有 Tc-99m-sestamibi 分布的扩展心脏躯干模型来生成包含呼吸和心脏运动的数据集,然后根据六个呼吸时相和八个心脏时相对它们进行分组。相应的衰减校正 (AC) 图用于模拟不同的 CT 协议:分别为双门控 CT (DCT)、呼吸门控 CT (RCT)、电影平均 CT (ACT)、插值 CT​​ (ICT)、吸气末 (HCT-in)、呼气末 (HCT-ex) 和呼吸中期 (HCT-mid) 的螺旋 CT。从 RAO 到 LPO 的 180° 范围内生成了 120 个无噪声且逼真的噪声投影,并通过 OS-EM 方法使用不同的 AC 贴图进行了重建。将不同呼吸时相的重建图像配准到一个参考呼气末时相,并将其汇总为每个心脏时相的心脏图像。生成相应的极坐标图并使用 17 段图进行进一步分析。使用以 DCT-AC 作为参考的极坐标图计算每个片段中平均强度的相对差 (RD)。在每个心动时相的 HCT-in AC 图像中都可以发现大量伪影。对于所有八个无噪声心脏时相,与 DCT 相比,ACT、HCT-ex、HCT-in、HCT-mid、RCT 和 ICT 的 AC 平均 RDmax 分别为 4.76%、4.95%、10.54%、4.23%、2.39% 和 2.76%。对于所有八个噪声心脏时相,与 DCT 相比,ACT、HCT-ex、HCT-in、HCT-mid、RCT 和 ICT 的 AC 的平均 RDmax 分别为 4.09%、5.04%、9.24%、4.02%、3.22% 和 2.85%。 ICT 和 RCT 对于双门控 SPECT 具有最佳的 AC 性能,而 HCT-ex、HCT-in、HCT-mid 和 ACT 显示较差的结果。由于复杂的实施和辐射问题,RCT 和 DCT 在临床中并不常用,因此建议使用 ICT,其次是 HCT-mid 或 HCT-ex,用于双门控心脏 SPECT 中的 AC,具有良好的准确性和相对较低的辐射剂量。
Dual respiratory-cardiac gating realizes motion-freeze myocardial perfusion single photon emission computed tomography (SPECT). The aim of this paper is to evaluate the performance of various attenuation maps for dual gating cardiac SPECT. We used extended cardiac-torso phantom with Tc-99m-sestamibi distribution to generate a dataset with both respiratory and cardiac motion and then grouped them based on six respiratory and eight cardiac phases. The corresponding attenuation correction (AC) maps were used to simulate different CT protocols: dual gating CT (DCT), respiratory gated CT (RCT), cine average CT (ACT), interpolated CT (ICT), helical CTs at end-inspiration (HCT-in), end-expiration (HCT-ex), and mid-respiration (HCT-mid), respectively. One hundred and twenty noise-free and realistic noisy projections were generated over 180° from RAO to LPO and reconstructed by OS-EM method using different AC maps. The reconstructed images in different respiratory phases were registered to one reference end-expiration phase and summed up as a cardiac image for each cardiac phase. The corresponding polar maps were generated and further analyzed using 17-segment plots. Relative difference (RD) of the average intensity was computed in each segment using the polar maps with DCT-AC as reference. Substantial artifacts could be found in HCT-in AC images in every cardiac phase. For all eight noise-free cardiac phases, the average RDmax for AC with ACT, HCT-ex, HCT-in, HCT-mid, RCT, and ICT comparing to DCT were 4.76%, 4.95%, 10.54%, 4.23%, 2.39%, and 2.76%, respectively. For all eight noisy cardiac phases, the average RDmax for AC with ACT, HCT-ex, HCT-in, HCT-mid, RCT, and ICT comparing to DCT were 4.09%, 5.04%, 9.24%, 4.02%, 3.22%, and 2.85%, respectively. The ICT and RCT have the best AC performance for dual gating SPECT, while HCT-ex, HCT-in, HCT-mid, and ACT show inferior results. Since RCT and DCT are not commonly used in clinics due to complex implementation and radiation concern, ICT is recommended, followed by HCT-mid or HCT-ex for AC in dual gating cardiac SPECT with good accuracy and relative low radiation dose.