Matsumoto Kazuhisa, Nakashima Masahiro, Kawai Tatsuya, Kawaguchi Takatsune, Sugihara Wataru, Urano Misugi, Nishikawa Yusuke, Kitera Nobuo, Watanabe Seita, Itoh Toshihide, Hiwatashi Akio
Department of Radiology, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya, Aichi 467-8601, Japan.
Department of Radiology, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya, Aichi 467-8601, Japan.
Eur J Radiol. 2025 Jul;188:112155. doi: 10.1016/j.ejrad.2025.112155. Epub 2025 May 7.
The middle meningeal artery (MMA) is the target arteries for the treatment of intracranial diseases. The purpose of this study was to investigate the ability of photon-counting detector CT (PCD-CT) to improve the visualization of the MMA compared to energy-integrated detector CT (EID-CT).
We retrospectively analyzed head CT angiography images from patients who underwent PCD-CT with ultra-high-resolution (UHR) mode between April 2023 and July 2024. We compared the images reconstructed with a slice thickness 0.2 mm and the matrix sizes 1024 × 1024 ('PCD-1024') and 512 × 512 ('PCD-512') with the same-patient images previously obtained by EID-CT with 0.75- or 1.0- mm slice thickness and the matrix size 512 × 512 ('EID-512'). Quantitatively, a region of interest (ROI) was placed on the maxillary artery near the origin of the MMA, and the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and contrast ratio (CR) were measured. Image sharpness was evaluated by measuring the full width at half maximum (FWHM) and the maximum intensity from the MMA vessel line profile. Qualitatively, the overall image quality, sharpness, and artifacts were evaluated using four-point Likert scales. The same evaluation was performed using PCD-CT with 0.2 mm ('0.2 mm') and virtual monoenergetic images (VMIs) ('45 keV', '55 keV', and '70 keV').
30 patients (19 males, 11 females; median age 65 years) were included. Quantitatively, the SNR (mean ± SD) for PCD-1024, PCD-512, and EID-512 were 16.9 ± 5.6, 20.3 ± 6.2, and 23.2 ± 7.8, respectively (P < 0.01 for PCD-1024 vs. EID-512). The CNR were 16.6 ± 4.5, 20.2 ± 5.5, and 44.4 ± 13.9. The CR were 0.74 ± 0.06, 0.74 ± 0.06, and 0.68 ± 0.08. The FWHM were 1.41 ± 0.17, 1.48 ± 0.15, and 1.77 ± 0.12, and the maximum intensity were 407.5 ± 71.7, 386.0 ± 72.5, and 218.2 ± 58.2 (P < 0.001 for PCD-1024 vs. EID-512 and PCD-512 vs. EID-512, respectively). Qualitatively, significant differences (P < 0.05) were identified among the PCD-1024, PCD-512, and EID-512 groups regarding the respective scores (median [IQR]) for overall image quality (4 [3-4], 3 [3-4], and 2 [2-3]), sharpness (4 [3-4], 3 [3-3], and 2 [2-3]), and artifacts (3 [3-3], 3 [2-3], and 2 [2-2]). In VMI analysis, quantitatively, 45 keV showed the highest values for SNR, CNR, CR, and maximum intensity; however, there were no differences in FWHM among the groups. Qualitatively, 0.2 mm was superior to the other groups in all evaluated aspects (P < 0.05).
PCD-1024 visualized the MMA better than both PCD-512 and EID-512. The 0.2 mm reconstruction was visually superior to the VMIs.
脑膜中动脉（MMA）是治疗颅内疾病的目标动脉。本研究的目的是探讨与能量积分探测器CT（EID-CT）相比，光子计数探测器CT（PCD-CT）改善MMA可视化的能力。
我们回顾性分析了2023年4月至2024年7月期间接受超高分辨率（UHR）模式PCD-CT检查的患者的头部CT血管造影图像。我们将重建的层厚为0.2mm、矩阵大小为1024×1024（“PCD-1024”）和512×512（“PCD-512”）的图像与之前通过层厚为0.75或1.0mm、矩阵大小为512×512（“EID-512”）的EID-CT获得的同患者图像进行比较。定量分析时，在MMA起源附近的上颌动脉上放置感兴趣区域（ROI），测量信噪比（SNR）、对比噪声比（CNR）和对比率（CR）。通过测量MMA血管轮廓线的半高宽（FWHM）和最大强度来评估图像清晰度。定性分析时，使用四点李克特量表评估整体图像质量、清晰度和伪影。使用层厚为0.2mm的PCD-CT（“0.2mm”）和虚拟单能图像（VMI）（“45keV”、“55keV”和“70keV”）进行相同的评估。
纳入30例患者（男性19例，女性11例；中位年龄65岁）。定量分析时，PCD-1024、PCD-512和EID-512的SNR（均值±标准差）分别为16.9±5.6、20.3±6.2和23.2±7.8（PCD-1024与EID-512相比，P<0.01）。CNR分别为16.6±4.5、20.2±5.5和44.4±13.9。CR分别为0.74±0.06、0.74±0.06和0.68±0.08。FWHM分别为1.41±0.17、1.48±0.15和1.77±0.12，最大强度分别为407.5±71.7、386.0±72.5和218.2±58.2（PCD-1024与EID-512以及PCD-512与EID-512相比，P均<0.001）。定性分析时，PCD-1024、PCD-512和EID-512组在整体图像质量（4[3-4]、3[3-4]和2[2-3]）、清晰度（4[3-4]、3[3-3]和2[2-3]）和伪影（3[3-3]、3[2-3]和2[2-2]）的各自评分（中位数[四分位间距]）方面存在显著差异（P<0.05）。在VMI分析中，定量分析时，45keV的SNR、CNR、CR和最大强度值最高；然而，各组之间FWHM无差异。定性分析时，0.2mm在所有评估方面均优于其他组（P<0.05）。
PCD-1024对MMA的可视化优于PCD-512和EID-512。0.2mm重建在视觉上优于VMI。
