Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada.
J Nucl Cardiol. 2009 Nov-Dec;16(6):914-26. doi: 10.1007/s12350-009-9132-7. Epub 2009 Aug 6.
Ongoing advancements in SPECT-CT technology raise important questions regarding the differences in performance between various cameras and their respective image-processing algorithms. Our study seeks the answer to this question via measurement of phantom myocardial wall thickness (MWT) on images obtained from three state-of-the-art cameras.
A thorax phantom with an insert modeling a healthy heart was scanned using cardiac acquisition protocols with Philips' Precedence (PP), GE's Infinia Hawkeye (IH), and Siemens' Symbia-T6 (SS). Processing was performed using advanced reconstruction techniques available on the cameras and our own independent software. The MWT measurement was used as a figure of merit in performance evaluation.
When using 50% threshold, MWTs measured for the data acquired using PP, IH, and SS and reconstructed with independent standardized software were 8.5 +/- 1.2 mm, 7.7 +/- 1.2 mm, and 9.3 +/- 0.9 mm, respectively; and 9.3 +/- 0.5 mm, 19.2 +/- 0.8 mm and 18.4 +/- 1.1 mm when using the manufacturers' own reconstructions, respectively. Thresholds optimized for each image (ranging from 32% to 68%) produced much more uniform results.
No significant differences were observed between image resolutions when data acquired from different cameras were reconstructed with an independent algorithm. However, different manufacturers' reconstruction algorithms produced MWTs that differed by up to about 110% when using a set threshold of 50%.
SPECT-CT 技术的不断进步提出了一个重要问题，即不同相机及其各自的图像处理算法在性能上存在差异。我们的研究通过测量来自三种最先进相机的图像中的幻影心肌壁厚度（MWT）来回答这个问题。
使用飞利浦的 Precedence（PP）、GE 的 Infinia Hawkeye（IH）和西门子的 Symbia-T6（SS）的心脏采集协议对带有插入物模拟健康心脏的胸腔模型进行扫描。使用相机上提供的高级重建技术和我们自己的独立软件进行处理。MWT 测量被用作性能评估的指标。
当使用 50%的阈值时，使用 PP、IH 和 SS 采集并使用独立标准化软件重建的数据的 MWT 分别为 8.5 +/- 1.2 毫米、7.7 +/- 1.2 毫米和 9.3 +/- 0.9 毫米；当使用制造商自己的重建时，分别为 9.3 +/- 0.5 毫米、19.2 +/- 0.8 毫米和 18.4 +/- 1.1 毫米。针对每个图像优化的阈值（从 32%到 68%）产生了更均匀的结果。
当使用独立算法重建来自不同相机的数据时，图像分辨率之间没有观察到显著差异。然而，当使用 50%的设定阈值时，不同制造商的重建算法产生的 MWT 差异最大可达 110%左右。
