文献速递技术与安全CT

深度学习时代CT图像质量评价:功能性与保真度之间的权衡

CT image quality evaluation in the age of deep learning: trade-off between functionality and fidelity.
PubMed 文献 · Eur Radiol · 2022-11-09

该研究在体模CT图像上定量比较DLIR与ASiR-V。DLIR在匹配噪声水平时噪声纹理更接近FBP、边缘增强更明显,但结构相似性低于ASiR-V,且不同解剖协议行为不同,临床部署需考虑这些权衡。

要点

  • DLIR噪声纹理更接近FBP,但结构相似性低于ASiR-V
  • DLIR较ASiR-V有额外边缘增强,阅片时注意过度锐化感
  • DLIR行为随扫描协议不同,临床参数设置需分部位优化

关键影像表现/征象

  • 腹部和胸部协议中,随着ASiR-V强度增加,NPS峰值频率从0.3 mm-1(FBP)降至0.15 mm-1(ASiR-V 100%),而所有DLIR级别的峰值频率保持在0.3 mm-1。
  • 脑协议中,随着DLIR级别增加,NPS峰值频率下移。
  • 三个DLIR级别的图像噪声水平分别与ASiR-V 40%、80%、100%相当。
  • 在匹配噪声水平时,DLIR的MSSIM低于ASiR-V,而MGM高于ASiR-V。

诊断标准与关键数值

  • 图像噪声与剂量呈强幂律关系(R2 > 0.99),适用于所有算法。
  • 定量指标包括噪声幅度σ、NPS峰值频率fpeak、平均SSIM(MSSIM)和平均梯度幅值(MGM)。
  • 以最大剂量FBP重建作为参考真值计算SSIM和差值图像梯度幅值。

临床应用与阅片要点

  • DLIR图像噪声纹理更接近FBP,但结构相似性较低,报告时需注意图像保真度可能下降。
  • DLIR较ASiR-V有额外边缘增强,阅片时警惕过度锐化伪影。
  • DLIR的噪声和纹理行为随扫描协议(脑、胸、腹)不同,临床参数应按部位分别优化。

局限性

  • 研究基于体模实验,非临床患者图像,结论外推需谨慎。
  • 仅评估了三种常规协议和特定剂量水平,未覆盖全部临床场景。

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: To quantitatively compare DLIR and ASiR-V with realistic anatomical images.

METHODS: CT scans of an anthropomorphic phantom were acquired using three routine protocols (brain, chest, and abdomen) at four dose levels, with images reconstructed at five levels of ASiR-V and three levels of DLIR. Noise power spectrum (NPS) was estimated using a difference image by subtracting two matching images from repeated scans. Using the max-dose FBP reconstruction as the ground truth, the structure similarity index (SSIM) and gradient magnitude (GM) of difference images were evaluated. Image noise magnitude (σ), frequency location of the NPS peak (fpeak), mean SSIM (MSSIM), and mean GM (MGM) were used as quantitative metrics to compare image quality, for each anatomical region, protocol, algorithm, dose level, and slice thickness.

RESULTS: Image noise had a strong (R2 > 0.99) power law relationship with dose for all algorithms. For the abdomen and chest, fpeak shifted from 0.3 (FBP) down to 0.15 mm-1 (ASiR-V 100%) with increasing ASiR-V strength but remained 0.3 mm-1 for all DLIR levels. fpeak shifted down for the brain protocol with increasing DLIR levels. Three levels of DLIR produced similar image noise levels as ASiR-V 40%, 80%, and 100%, respectively. DLIR had lower MSSIM but higher MGM than ASiR-V while matching imaging noise.

CONCLUSION: Compared to ASiR-V, DLIR presents trade-offs between functionality and fidelity: it has a noise texture closer to FBP and more edge enhancement, but reduced structure similarity. These trade-offs and unique protocol-dependent behaviors of DLIR should be considered during clinical implementation and deployment.

KEY POINTS: • DLIR reconstructed images demonstrate closer noise texture and lower structure similarity to FBP while producing equivalent noise levels comparable to ASiR-V. • DLIR has additional edge enhancement as compared to ASiR-V. • DLIR has unique protocol-dependent behaviors that should be considered for clinical implementation.

来源:Eur Radiol,查看原文。版权归原作者所有。

引用

  1. Yang, Cao, Pisuchpen, et al. CT image quality evaluation in the age of deep learning: trade-off between functionality and fidelity. Eur Radiol. 2022. PMID: 36350391. doi:10.1007/s00330-022-09233-0

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。