文献速递头颈神经CT

光子计数探测器CT与能量积分探测器CT用于颞骨成像的比较:一项尸体研究

Comparison of a Photon-Counting-Detector CT with an Energy-Integrating-Detector CT for Temporal Bone Imaging: A Cadaveric Study.
PubMed 文献 · AJNR Am J Neuroradiol · 2018-08-09

该尸体研究比较了光子计数探测器CT与能量积分探测器CT的超高分辨率模式对颞骨精细结构的显示能力。体模实验显示光子计数探测器CT面内分辨率提高,相同0.6毫米层厚下图像噪声显著更低(P<.001)。3名神经放射科医生盲法评估10例尸体颞骨,对砧锤关节、镫骨足板、蜗轴及整体图像质量均更偏好光子计数探测器CT图像,读者间一致性中等(Krippendorff α=0.50)。

要点

  • 光子计数探测器CT在相同0.6毫米层厚下图像噪声显著低于能量积分探测器CT(P<.001)。
  • 读者对砧锤关节、镫骨足板、蜗轴及整体图像质量的排名均显著偏好光子计数探测器CT(P<.001)。
  • 读者间一致性为中等(Krippendorff α=0.50)。

核心结论

  • 光子计数探测器CT的超高分辨率模式在显示颞骨精细解剖方面明显优于商用能量积分探测器CT的超高分辨率模式。

研究设计与样本

  • 尸体研究,采用10例尸体颞骨,在原型光子计数探测器CT和带梳状滤波器的能量积分探测器CT上扫描,辐射剂量与临床检查匹配;图像重建采用锐利核,能量积分探测器CT层厚0.6毫米(最小),光子计数探测器CT层厚0.6毫米和0.25毫米(最小);体模研究量化空间分辨率;3名神经放射科医生盲法评估。

关键影像表现/征象

  • 光子计数探测器CT较能量积分探测器CT面内分辨率提高。
  • 相同0.6毫米层厚下,光子计数探测器CT图像噪声显著低于能量积分探测器CT(P<.001)。
  • 读者对砧锤关节、镫骨足板、蜗轴3个结构的图像偏好均倾向于光子计数探测器CT。
  • 整体图像质量方面,光子计数探测器CT图像集排名显著高于能量积分探测器CT(P<.001)。

诊断标准与关键数值

  • 读者间一致性Krippendorff α=0.50,为中等一致性。
  • 图像噪声比较采用校正的单因素方差分析,噪声差异P<.001。
  • 各标本和各方案的排名比较采用Friedman检验,整体图像质量排名差异P<.001。

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)。

英文原文(摘要)· 点击展开

BACKGROUND AND PURPOSE: Evaluating abnormalities of the temporal bone requires high-spatial-resolution CT imaging. Our aim was to assess the performance of photon-counting-detector ultra-high-resolution acquisitions for temporal bone imaging and compare the results with those of energy-integrating-detector ultra-high-resolution acquisitions.

MATERIALS AND METHODS: Phantom studies were conducted to quantify spatial resolution of the ultra-high-resolution mode on a prototype photon-counting-detector CT scanner and an energy-integrating-detector CT scanner that uses a comb filter. Ten cadaveric temporal bones were scanned on both systems with the radiation dose matched to that of the clinical examinations. Images were reconstructed using a sharp kernel, 0.6-mm (minimum) thickness for energy-integrating-detector CT, and 0.6- and 0.25-mm (minimum) thicknesses for photon-counting-detector CT. Image noise was measured and compared using adjusted 1-way ANOVA. Images were reviewed blindly by 3 neuroradiologists to assess the incudomallear joint, stapes footplate, modiolus, and overall image quality. The ranking results for each specimen and protocol were compared using the Friedman test. The Krippendorff α was used for interreader agreement.

RESULTS: Photon-counting-detector CT showed an increase of in-plane resolution compared with energy-integrating-detector CT. At the same thickness (0.6 mm), images from photon-counting-detector CT had significantly lower (P < .001) image noise compared with energy-integrating-detector CT. Readers preferred the photon-counting-detector CT images to the energy-integrating-detector images for all 3 temporal bone structures. A moderate interreader agreement was observed with the Krippendorff α = 0.50. For overall image quality, photon-counting-detector CT image sets were ranked significantly higher than images from energy-integrating-detector CT (P < .001).

CONCLUSIONS: This study demonstrated substantially better delineation of fine anatomy for the temporal bones scanned with the ultra-high-resolution mode of photon-counting-detector CT compared with the ultra-high-resolution mode of a commercial energy-integrating-detector CT scanner.

来源:AJNR Am J Neuroradiol,查看原文。版权归原作者所有。

引用

  1. Zhou W, Lane JI, Carlson ML, et al. Comparison of a Photon-Counting-Detector CT with an Energy-Integrating-Detector CT for Temporal Bone Imaging: A Cadaveric Study. AJNR Am J Neuroradiol. 2018. PMID: 30093479. doi:10.3174/ajnr.A5768

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。