肩部T2及中间加权MRI高斯混合模型肌内脂肪定量
结果显示GMM在T2W和IW上均优于K-means,T2W上多数肌肉可靠性极佳,且GMM是唯一与Dixon无显著差异的模型。
要点
- GMM在T2W肩部MRI上对冈上肌、肩胛下肌、冈下肌及三角肌的IMF定量可靠性极佳。
- GMM在T2W和IW MRI上与Dixon IMF值均无显著差异,可用于替代脂肪定量。
- K-means在肩部MRI肌内脂肪定量中偏倚较大、可靠性较差,不宜常规使用。
核心结论
- GMM阈值法在T2W和IW肩部MRI上对肌内脂肪的定量与Dixon的一致性优于K-means。
诊断标准与关键数值
- T2W MRI上GMM最大平均偏倚为-3.07%,冈上肌、肩胛下肌、冈下肌及三角肌ICC2,1≥0.90。
- IW MRI上GMM平均偏倚不超过-2.81%,冈下肌ICC2,1=0.92,其余肌肉可靠性良好。
- K-means在IW MRI上偏倚为-4.36%,ICC2,1≤0.81。
- GMM是唯一在T2W(p≥0.074)和IW(p≥0.243)MRI上与DixonIMF值无显著差异的模型。
临床应用与阅片要点
- 在无Dixon序列的T2W或IW肩部MRI上,可考虑用GMM自动阈值法进行肌内脂肪定量替代评估。
- 阅片时如需报告肩袖肌脂肪浸润程度,GMM结果与Dixon一致性更好,优于K-means聚类。
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 accuracy and reliability of automated Gaussian Mixture Modelling (GMM) and K-means clustering for intramuscular fat (IMF) quantification on T2-weighted (T2W) and intermediate-weighted (IW) shoulder MRI, using two-point Dixon as the reference standard.
MATERIALS & METHODS: This retrospective study collected MRI datasets of patients who underwent both two-point Dixon with T2W and/or IW shoulder MRI. GMM and K-means were applied to quantify IMF of the supraspinatus, subscapularis, infraspinatus, teres minor, and deltoid. Accuracy and reliability relative to Dixon were evaluated using Bland-Altman analysis and intraclass correlation coefficients (ICC2,1). Raw IMF values from GMM, K-means and Dixon were compared using repeated-measures ANOVA with post-hoc Dunnett's test (∝=0.05).
RESULTS: Eighty patients (mean age, 58 years ± 15 [standard deviation]; 41 males) were evaluated. On T2W MRI, GMM outperformed K-means with a maximum mean bias of -3.07% and excellent reliability for supraspinatus, subscapularis, infraspinatus and deltoid (ICC2,1 ≥ 0.90). K-means produced poor to moderate reliability. On IW MRI, GMM mean bias did not exceed -2.81% and showed excellent reliability for infraspinatus (ICC2,1 = 0.92) with good reliability for other muscles. K-means showed higher bias (-4.36%) and lower reliability (ICC2,1 ≤ 0.81) than GMM. GMM was the only model that did not significantly differ from Dixon IMF values on T2W (p ≥ 0.074) and IW (p ≥ 0.243) MRI.
CONCLUSION: IMF quantification using GMM thresholding achieves better agreement with Dixon IMF values than K-Means on T2W and IW MRI.
引用
- Kim, Wesselink, Gandomkar, et al. Gaussian mixture modelling enables valid intramuscular fat quantification on T2- and intermediate-weighted shoulder MRI relative to two-point Dixon. Eur J Radiol. 2026. PMID: 42710425. doi:10.1016/j.ejrad.2026.113206
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。