文献速递骨肌MR

深度学习重建的3D MRI用于膝关节综合评估:与1.5 T多序列2D方案的比较

Deep-learning reconstructed 3D MRI for comprehensive knee assessment: Comparison with a multisequence 2D protocol at 1.5 T.
PubMed 文献 · Eur J Radiol · 2025-09-27

回顾性纳入95例成人膝关节1.5 T MRI,比较单一各向同性3D PD加权脂肪饱和序列与五序列2D方案(均用深度学习重建)。多数结构评估无显著差异,但外侧软骨、外侧骨髓水肿和滑膜炎评估有显著差异,图像质量与伪影无显著差异。

要点

  • 3D与2D方案在半月板、内侧及髌股软骨、韧带、骨折等多数结构评估无显著差异。
  • 外侧软骨、外侧骨髓水肿和滑膜炎评估两方案有显著差异。
  • 图像质量与伪影无显著差异,读者间一致性均为中等程度。

核心结论

  • 单一3D-PDFS与多序列2D方案(均用深度学习重建)在膝关节评估中表现多数可比,但少数结构存在差异。

研究设计与样本

  • 回顾性收集2023年5月至2024年7月连续成人膝关节1.5 T MRI病例95例,女性占39%,年龄52.7±14.5岁。

诊断标准与关键数值

  • 采用Likert量表和二分类分级评估内侧、外侧及髌股间室的异常、整体图像质量、诊断信心和伪影。

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 compare a single isotropic 3D PD-weighted fat-saturated (3D-PDFS) acquisition with a standard 2D multisequence protocol in knee MRI using deep learning reconstruction (DLR) for comprehensive assessment of knee structures.

MATERIALS AND METHODS: We retrospectively collected consecutive adult patients undergoing 1.5 T knee MRI between May 2023 and July 2024. The MRI protocol comprised one 3D-PDFS and five multiplanar 2D sequences with DLR. Two radiologists independently evaluated the 3D and 2D sequences, assessing abnormalities for medial (MC), lateral (LC), and patellofemoral (PF) compartments, overall image quality, diagnostic confidence, and artifacts. Likert-scale and binary grades were compared via Wilcoxon signed-rank and McNemar tests, respectively. Inter-reader agreement was assessed via Cohen's kappa.

RESULTS: A total of 95 patients (females, 39 %; age, 52.7 ± 14.5 y) were included. The 3D and 2D protocols did not significantly differ regarding the assessment of menisci, MC and PF cartilage, collateral ligaments, cruciate ligaments, MC and PF bone marrow edema (BME), and fractures (p > 0.057). Significant differences were shown in assessment of LC cartilage (p = 0.002), LC BME (p = 0.040) and synovitis (p = 0.002). Image quality and artifacts were not significantly different (p > 0.082). Diagnostic confidence of one reader was significantly higher for the 2D protocol (p = 0.023). Average inter-reader agreement was substantial for the 3D (k = 0.67) and 2D (k = 0.66) protocols.

CONCLUSION: Our results suggest mostly comparable performance between a single 3D-PDFS and a multisequence 2D protocol using DLR for knee assessment, with the exception of few structures.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Nikolova E, Kroschke J, Obermüller C, et al. Deep-learning reconstructed 3D MRI for comprehensive knee assessment: Comparison with a multisequence 2D protocol at 1.5 T. Eur J Radiol. 2025. PMID: 41045731. doi:10.1016/j.ejrad.2025.112458

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。