Priya Sarv, Dhruba Durjoy D, Sorensen Eldon, Aher Pritish Y, Narayanasamy Sabarish, Nagpal Prashant, Jacob Mathews, Carter Knute D
From the Department of Radiology, University of Iowa Carver College of Medicine, 200 Hawkins Dr, Iowa City, IA 52242 (S.P., S.N.); Department of Electrical and Computer Engineering (D.D.D., M.J.) and Department of Biostatistics (E.S., K.D.C.), University of Iowa, Iowa City, Iowa; Department of Radiology, University of Miami, Miller School of Medicine, Miami, Fla (P.Y.A.); and Department of Radiology, University of Wisconsin School of Medicine and Public Health, Madison, Wis (P.N.).
Radiol Cardiothorac Imaging. 2023 Jul 27;5(4):e220312. doi: 10.1148/ryct.220312. eCollection 2023 Aug.
To investigate the effect of ComBat harmonization methods on the robustness of cardiac MRI-derived radiomic features to variations in imaging parameters.
This Health Insurance Portability and Accountability Act-compliant retrospective study used a publicly available data set of 11 healthy controls (mean age, 33 years ± 16 [SD]; six men) and five patients (mean age, 52 years ± 16; four men). A single midventricular short-axis section was acquired with 3-T MRI using cine balanced steady-state free precision, T1-weighted, T2-weighted, T1 mapping, and T2 mapping imaging sequences. Each sequence was acquired using baseline parameters and after variations in flip angle, spatial resolution, section thickness, and parallel imaging. Image registration was performed for all sequences at a per-individual level. Manual myocardial contouring was performed, and 1652 radiomic features per sequence were extracted using baseline and variations in imaging parameters. Radiomic feature stability to change in imaging parameters was assessed using Cohen sensitivity. The stability of radiomic features was assessed both without and after ComBat harmonization of radiomic features. Three ComBat methods were studied: parametric, nonparametric, and Gaussian mixture model (GMM).
For all sequences combined, 51.4% of features were robust to changes in imaging parameters when no ComBat method was applied. ComBat harmonization substantially increased the number of stable features to 95.1% (95% CI: 94.9, 95.3) when parametric ComBat was used and 90.9% (95% CI: 90.6, 91.2) when nonparametric ComBat was used. GMM combat resulted in only 52.6% stable features.
ComBat harmonization improved the stability of radiomic features to changes in imaging parameters across all cardiac MRI sequences. Cardiac MRI, Radiomics, ComBat, Harmonization © RSNA, 2023.
研究ComBat归一化方法对心脏磁共振成像衍生的放射组学特征针对成像参数变化的稳健性的影响。
这项符合《健康保险流通与责任法案》的回顾性研究使用了一个公开可用的数据集，其中包括11名健康对照者（平均年龄33岁±16[标准差]；6名男性）和5名患者（平均年龄52岁±16；4名男性）。使用3-T磁共振成像通过电影平衡稳态自由精度、T1加权、T2加权、T1映射和T2映射成像序列获取单个心室中部短轴切面。每个序列均使用基线参数以及在翻转角、空间分辨率、层厚和平行成像发生变化后进行采集。在个体水平上对所有序列进行图像配准。进行手动心肌轮廓勾画，并使用基线和成像参数变化提取每个序列的1652个放射组学特征。使用科恩敏感性评估放射组学特征对成像参数变化的稳定性。在对放射组学特征进行ComBat归一化之前和之后均评估放射组学特征的稳定性。研究了三种ComBat方法：参数法、非参数法和高斯混合模型（GMM）。
对于所有组合序列，在未应用ComBat方法时，51.4%的特征对成像参数变化具有稳健性。当使用参数ComBat时，ComBat归一化将稳定特征的数量大幅增加至95.1%（95%置信区间：94.9，95.3），当使用非参数ComBat时增加至90.9%（95%置信区间：90.6，91.2）。GMM-ComBat仅产生52.6%的稳定特征。
ComBat归一化提高了所有心脏磁共振成像序列中放射组学特征针对成像参数变化的稳定性。心脏磁共振成像、放射组学、ComBat、归一化 ©RSNA，2023年
