文献速递腹部技术与安全CT

超高分辨率光子计数CT胰腺成像:图像质量初步评估

Pancreatic imaging with ultra-high-resolution photon-counting computed tomography: Preliminary assessment of image quality.
PubMed 文献 · Eur J Radiol · 2025-12-01

前瞻性纳入74例组织学确诊胰腺导管腺癌患者,行超高分辨率模式胰腺多期增强CT,重建6组图像系列。结果显示0.4 mm层厚联合低 keV 虚拟单能图像(尤其45 keV)在肿瘤和胰周血管显示及精细解剖细节上图像质量更优。

要点

  • 0.4 mm层厚联合45 keV虚拟单能图像在胰腺期和门静脉期对肿瘤及胰周血管显示质量最佳。
  • 肿瘤和胰周血管的CNR与SNR随虚拟单能级升高而下降,在40 keV达峰值(SNRlesion除外)。
  • 1 mm层厚70 keV图像噪声最低,但精细解剖细节评分低于0.4 mm 45 keV和50 keV图像。

核心结论

  • 在光子计数CT中,0.4 mm层厚联合低 keV 虚拟单能图像重建(尤其45 keV)为胰腺导管腺癌术前评估提供更好的图像质量。

研究设计与样本

  • 前瞻性研究,2024年9月至2025年4月纳入74例疑似胰腺导管腺癌并最终组织学确诊的患者(39例男性,平均年龄65.7±10.7岁),采用超高分辨率扫描模式行胰腺多期增强CT,重建6组图像系列。

关键影像表现/征象

  • 肿瘤和胰周血管的CNR与SNR在胰腺实质期和门静脉期随虚拟单能级升高而下降,40 keV达峰值(所有P<0.001,SNRlesion除外)。
  • 图像噪声在1 mm层厚70 keV系列最低(所有P<0.001)。

诊断标准与关键数值

  • 定性评估采用5点Likert量表,由两名盲法观察者评价肿瘤、胰周血管及精细胰腺解剖的图像质量。
  • 整体图像质量在各期均以1 mm-70 keV、0.4 mm-45 keV和0.4 mm-50 keV图像更优(所有P<0.05),但门静脉期1 mm-70 keV与0.4 mm-55 keV图像间除外。
  • 精细解剖细节评分以0.4 mm-45 keV和0.4 mm-50 keV图像高于其他系列。

临床应用与阅片要点

  • 在光子计数CT胰腺多期增强扫描中,可考虑采用0.4 mm层厚联合45 keV虚拟单能图像重建,以优化胰腺导管腺癌及胰周血管的显示。

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

OBJECTIVES: To evaluate the advantage of combining ultra-high-resolution (UHR) mode with virtual monoenergetic image (VMI) reconstruction in photon-counting CT (PCCT) to assess pancreatic ductal adenocarcinoma (PDAC) and peripancreatic vasculature.

METHODS: From September 2024 to April 2025, participants with suspected PDAC were prospectively enrolled and underwent pancreatic multiphase contrast-enhanced CT in UHR scanning mode. Six image series were reconstructed: (a) 0.2-mm slice thickness, threshold three-dimensional images; (b) 1-mm slice thickness, VMIs at 70 keV; and (c-f) 0.4-mm slice thickness, VMIs at 40, 45, 50, and 55 keV. For quantitative analysis, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were measured in tumors and peripancreatic vessels. For qualitative analysis, two blinded observers evaluated image quality for tumors, peripancreatic vessels, and fine pancreatic anatomy using 5-point Likert-type scales.

RESULTS: Seventy-four participants (39 male; mean age, 65.7 ± 10.7 years) with histologically confirmed PDAC were included. The CNR and SNR for tumors and peripancreatic vessels declined with increasing monoenergetic levels during pancreatic parenchymal and portal venous phases (PVP), peaking at 40 keV (all P < 0.001, except for SNRlesion). Image noise was lowest in 1-mm-70-keV series (all P < 0.001). Qualitative assessment favored 1-mm-70-keV, 0.4-mm-45-keV, and 0.4-mm-50-keV images for overall image quality across phases (all P < 0.05), except between 1-mm-70-keV and 0.4-mm-55-keV images during PVP. For fine anatomical details, 0.4-mm-45-keV and 0.4-mm-50-keV images received higher ratings relative to other series.

CONCLUSION: In PCCT, combining 0.4-mm slice thickness with low-keV VMI reconstruction (particularly at 45 keV) provides better image quality for preoperative evaluation of PDAC.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Huang R, Yang Y, Gao M, et al. Pancreatic imaging with ultra-high-resolution photon-counting computed tomography: Preliminary assessment of image quality. Eur J Radiol. 2025. PMID: 41352230. doi:10.1016/j.ejrad.2025.112596

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。