文献速递胸部CT

验证深度学习图像重建用于低剂量胸部CT扫描:聚焦图像质量与噪声

Validation of Deep-Learning Image Reconstruction for Low-Dose Chest Computed Tomography Scan: Emphasis on Image Quality and Noise.
PubMed 文献 · Korean J Radiol · 2020-07-27

该回顾性研究纳入58例肺癌筛查低剂量胸部CT患者,比较深度学习图像重建(DLIR)中、高等级与自适应统计迭代重建-Veo 30%(ASiR-V 30%)的图像质量与噪声。客观分析显示DLIR-M和DLIR-H噪声显著低于ASiR-V 30%(均P<0.001),信噪比和对比噪声比更高;主观分析显示DLIR图像对比度更高、噪声更低,结构辨识度优于ASiR-V 30%。结论认为DLIR在维持更优图像质量的同时显著降低图像噪声。

要点

  • DLIR-M和DLIR-H的图像噪声显著低于ASiR-V 30%(均P<0.001)。
  • DLIR的SNR和CNR均高于ASiR-V 30%,肺、纵隔、肝脏SNR的P值分别为0.027、<0.001、<0.001。
  • DLIR在识别肺动静脉、气管支气管、淋巴结、胸膜和心包方面优于ASiR-V 30%(均P<0.001)。

核心结论

  • 与ASiR-V 30%相比,DLIR显著降低胸部低剂量CT图像的噪声,同时维持更优的图像质量。

研究设计与样本

  • 回顾性研究,纳入58例接受低剂量胸部CT肺癌筛查的患者。
  • 数据集分别用ASiR-V 30%、DLIR-M和DLIR-H重建。
  • 客观评价肺、纵隔、肝脏和背景空气的平均衰减值(HU)和标准差(噪声);主观评价图像对比度、噪声和结构辨识度。

关键影像表现/征象

  • 客观图像信号在ASiR-V 30%、DLIR-M和DLIR-H之间无显著差异(肺、纵隔、肝脏和背景空气的P值分别为0.949、0.737、0.366和0.358)。
  • DLIR-M和DLIR-H的噪声显著低于ASiR-V 30%(均P<0.001)。
  • DLIR的SNR和CNR高于ASiR-V 30%(肺、纵隔、肝脏SNR的P值分别为0.027、<0.001、<0.001;CNR均P<0.001)。
  • 主观分析显示DLIR图像对比度更高、噪声更低(均P<0.001)。
  • DLIR在识别肺动静脉、气管支气管、淋巴结、胸膜和心包方面优于ASiR-V 30%(均P<0.001)。

临床应用与阅片要点

  • 在低剂量胸部CT肺癌筛查中,DLIR-M和DLIR-H可在不改变图像信号的情况下降低噪声并提高SNR和CNR。
  • DLIR有助于改善肺动静脉、气管支气管、淋巴结、胸膜和心包的辨识度。

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

OBJECTIVE: Iterative reconstruction degrades image quality. Thus, further advances in image reconstruction are necessary to overcome some limitations of this technique in low-dose computed tomography (LDCT) scan of the chest. Deep-learning image reconstruction (DLIR) is a new method used to reduce dose while maintaining image quality. The purposes of this study was to evaluate image quality and noise of LDCT scan images reconstructed with DLIR and compare with those of images reconstructed with the adaptive statistical iterative reconstruction-Veo at a level of 30% (ASiR-V 30%).

MATERIALS AND METHODS: This retrospective study included 58 patients who underwent LDCT scan for lung cancer screening. Datasets were reconstructed with ASiR-V 30% and DLIR at medium and high levels (DLIR-M and DLIR-H, respectively). The objective image signal and noise, which represented mean attenuation value and standard deviation in Hounsfield units for the lungs, mediastinum, liver, and background air, and subjective image contrast, image noise, and conspicuity of structures were evaluated. The differences between CT scan images subjected to ASiR-V 30%, DLIR-M, and DLIR-H were evaluated.

RESULTS: Based on the objective analysis, the image signals did not significantly differ among ASiR-V 30%, DLIR-M, and DLIR-H (p = 0.949, 0.737, 0.366, and 0.358 in the lungs, mediastinum, liver, and background air, respectively). However, the noise was significantly lower in DLIR-M and DLIR-H than in ASiR-V 30% (all p < 0.001). DLIR had higher signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) than ASiR-V 30% (p = 0.027, < 0.001, and < 0.001 in the SNR of the lungs, mediastinum, and liver, respectively; all p < 0.001 in the CNR). According to the subjective analysis, DLIR had higher image contrast and lower image noise than ASiR-V 30% (all p < 0.001). DLIR was superior to ASiR-V 30% in identifying the pulmonary arteries and veins, trachea and bronchi, lymph nodes, and pleura and pericardium (all p < 0.001).

CONCLUSION: DLIR significantly reduced the image noise in chest LDCT scan images compared with ASiR-V 30% while maintaining superior image quality.

来源:Korean J Radiol,查看原文。版权归原作者所有。

引用

  1. Kim JH, Yoon HJ, Lee E, et al. Validation of Deep-Learning Image Reconstruction for Low-Dose Chest Computed Tomography Scan: Emphasis on Image Quality and Noise. Korean J Radiol. 2020. PMID: 32729277. doi:10.3348/kjr.2020.0116

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。