Barbieri Marco, Gatti Anthony A, Kogan Feliks
Department of Radiology, Stanford University, Stanford, California, USA.
J Magn Reson Imaging. 2025 May;61(5):2116-2127. doi: 10.1002/jmri.29646. Epub 2024 Oct 28.
The Osteoarthritis Initiative (OAI) collected extensive imaging data, including Multi-Echo Spin-Echo (MESE) sequences for measuring knee cartilage T relaxation times. Mono-exponential models are used in the OAI for T fitting, which neglects stimulated echoes and B inhomogeneities. Extended Phase Graph (EPG) modeling addresses these limitations but has not been applied to the OAI dataset.
To assess how different fitting methods, including EPG-based and exponential-based approaches, affect the accuracy and reproducibility of cartilage T in the OAI dataset.
Retrospective.
From OAI dataset, 50 subjects, stratified by osteoarthritis (OA) severity using Kellgren-Lawrence grades (KLG), and 50 subjects without OA diagnosis during OAI duration were selected (each group: 25 females, mean ages ~61 years).
FIELD STRENGTH/SEQUENCE: 3-T, two-dimensional (2D) MESE sequence.
Femoral and tibial cartilages were segmented from DESS images, subdivided into seven sub-regions, and co-registered to MESE. T maps were obtained using three EPG-based methods (nonlinear least squares, dictionary matching, and deep learning) and three mono-exponential approaches (linear least squares, nonlinear least squares, and noise-corrected exponential). Average T values within sub-regions were obtained. Pair-wise agreement among fitting methods was evaluated using the stratified subjects, while reproducibility using healthy subjects. Each method's T accuracy and repeatability varying signal-to-noise ratio (SNR) were assessed with simulations.
Bland-Altman analysis, Lin's concordance coefficient, and coefficient of variation assessed agreement, repeatability, and reproducibility. Statistical significance was set at P-value <0.05.
EPG-based methods demonstrated superior T accuracy (mean absolute error below 0.5 msec at SNR > 100) compared to mono-exponential methods (error > 7 msec). EPG-based approaches had better reproducibility, with limits of agreement 1.5-5 msec narrower than exponential-based methods. T values from EPG methods were systematically 10-17 msec lower than those from mono-exponential fitting.
EPG modeling improved agreement and reproducibility of cartilage T mapping in subjects from the OAI dataset.
3 TECHNICAL EFFICACY: Stage 1.
骨关节炎倡议（OAI）收集了大量成像数据，包括用于测量膝关节软骨T2弛豫时间的多回波自旋回波（MESE）序列。OAI中使用单指数模型进行T2拟合，该模型忽略了受激回波和B场不均匀性。扩展相位图（EPG）建模解决了这些局限性，但尚未应用于OAI数据集。
评估不同的拟合方法，包括基于EPG和基于指数的方法，如何影响OAI数据集中软骨T2的准确性和可重复性。
回顾性研究。
从OAI数据集中，选择50名受试者，根据Kellgren-Lawrence分级（KLG）按骨关节炎（OA）严重程度分层，以及50名在OAI期间未诊断为OA的受试者（每组：25名女性，平均年龄约61岁）。
场强/序列：3-T，二维（2D）MESE序列。
从DESS图像中分割出股骨和胫骨软骨，细分为七个子区域，并与MESE图像进行配准。使用三种基于EPG的方法（非线性最小二乘法、字典匹配法和深度学习法）和三种单指数方法（线性最小二乘法、非线性最小二乘法和噪声校正指数法）获得T2图。获取子区域内的平均T2值。使用分层受试者评估拟合方法之间的成对一致性，使用健康受试者评估可重复性。通过模拟评估每种方法在不同信噪比（SNR）下的T2准确性和重复性。
Bland-Altman分析、Lin一致性系数和变异系数评估一致性、重复性和可重复性。统计学显著性设定为P值<0.05。
与单指数方法（误差>7毫秒）相比，基于EPG的方法显示出更高的T2准确性（在SNR>100时平均绝对误差低于0.5毫秒）。基于EPG的方法具有更好的可重复性，一致性界限比基于指数的方法窄1.5-5毫秒。EPG方法得到的T2值比单指数拟合得到的T2值系统地低10-17毫秒。
EPG建模提高了OAI数据集中受试者软骨T2映射的一致性和可重复性。
3级 技术效能：1级
