From the Departments of Radiology (N.C.D.T., C.G.B.Y., B.C.W., J.M.H., D.R., N.S., B.F., M.C.P., A.J.M., J.A.M.), Pathology (K.J.H.), and Neurologic Surgery (T.R.P.), The University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390; Department of Bioengineering, The University of Texas at Dallas, Richardson, Tex (B.F.); Departments of Radiology (M.D.L., R.J.) and Neurosurgery (R.J.), New York University Grossman School of Medicine, New York, NY; and Department of Radiology, University of Wisconsin-Madison, Madison, Wis (R.J.B.).
Radiol Artif Intell. 2024 Jul;6(4):e230218. doi: 10.1148/ryai.230218.
Purpose To develop a radiomics framework for preoperative MRI-based prediction of isocitrate dehydrogenase () mutation status, a crucial glioma prognostic indicator. Materials and Methods Radiomics features (shape, first-order statistics, and texture) were extracted from the whole tumor or the combination of nonenhancing, necrosis, and edema regions. Segmentation masks were obtained via the federated tumor segmentation tool or the original data source. Boruta, a wrapper-based feature selection algorithm, identified relevant features. Addressing the imbalance between mutated and wild-type cases, multiple prediction models were trained on balanced data subsets using random forest or XGBoost and assembled to build the final classifier. The framework was evaluated using retrospective MRI scans from three public datasets (The Cancer Imaging Archive [TCIA, 227 patients], the University of California San Francisco Preoperative Diffuse Glioma MRI dataset [UCSF, 495 patients], and the Erasmus Glioma Database [EGD, 456 patients]) and internal datasets collected from the University of Texas Southwestern Medical Center (UTSW, 356 patients), New York University (NYU, 136 patients), and University of Wisconsin-Madison (UWM, 174 patients). TCIA and UTSW served as separate training sets, while the remaining data constituted the test set (1617 or 1488 testing cases, respectively). Results The best performing models trained on the TCIA dataset achieved area under the receiver operating characteristic curve (AUC) values of 0.89 for UTSW, 0.86 for NYU, 0.93 for UWM, 0.94 for UCSF, and 0.88 for EGD test sets. The best performing models trained on the UTSW dataset achieved slightly higher AUCs: 0.92 for TCIA, 0.88 for NYU, 0.96 for UWM, 0.93 for UCSF, and 0.90 for EGD. Conclusion This MRI radiomics-based framework shows promise for accurate preoperative prediction of mutation status in patients with glioma. Glioma, Isocitrate Dehydrogenase Mutation, Mutation, Radiomics, MRI Published under a CC BY 4.0 license. See also commentary by Moassefi and Erickson in this issue.
目的 开发一种基于术前磁共振成像（MRI）的放射组学框架，用于预测异柠檬酸脱氢酶（IDH）突变状态，这是一种关键的胶质瘤预后标志物。
材料与方法 从整个肿瘤或非增强区、坏死区和水肿区的组合中提取放射组学特征（形状、一阶统计量和纹理）。分割掩模通过联邦肿瘤分割工具或原始数据源获得。使用基于包装的特征选择算法 Boruta ，筛选出相关特征。为了解决突变型和野生型病例之间的不平衡问题，使用随机森林或 XGBoost 在平衡数据子集中训练多个预测模型，并将其组合起来构建最终分类器。该框架使用来自三个公共数据集（癌症成像档案 [TCIA，227 例]、加州大学旧金山分校术前弥漫性胶质瘤 MRI 数据集 [UCSF，495 例]和 Erasmus 神经胶质瘤数据库 [EGD，456 例]）和来自德克萨斯大学西南医学中心（UTSW，356 例）、纽约大学（NYU，136 例）和威斯康星大学麦迪逊分校（UWM，174 例）的内部数据集进行了评估。TCIA 和 UTSW 作为独立的训练集，其余数据构成了测试集（分别为 1617 或 1488 个测试病例）。
结果 在 TCIA 数据集上训练的表现最佳的模型在 UTSW、NYU、UWM、UCSF 和 EGD 测试集的受试者工作特征曲线下面积（AUC）值分别为 0.89、0.86、0.93、0.94 和 0.88。在 UTSW 数据集上训练的表现最佳的模型的 AUC 值略高：TCIA 为 0.92、NYU 为 0.88、UWM 为 0.96、UCSF 为 0.93、EGD 为 0.90。
结论 该基于 MRI 放射组学的框架有望实现胶质瘤患者术前 IDH 突变状态的准确预测。
关键词：胶质瘤，异柠檬酸脱氢酶突变，突变，放射组学，MRI
发表于 CC BY 4.0 许可下。参见本期 Moassefi 和 Erickson 的评论。
Zotero 插件
