PET-CT/MR Department, Harbin Medical University Cancer Hospital, Harbin, China.
Medical Imaging Department, The Fourth Affiliated Hospital of Harbin Medical University, Harbin, China.
Neuroradiology. 2022 Jul;64(7):1373-1382. doi: 10.1007/s00234-022-02894-0. Epub 2022 Jan 17.
This study aimed to investigate the clinical usefulness of the enhanced-T1WI-based deep learning radiomics model (DLRM) in differentiating low- and high-grade meningiomas.
A total of 132 patients with pathologically confirmed meningiomas were consecutively enrolled (105 in the training cohort and 27 in the test cohort). Radiomics features and deep learning features were extracted from T1 weighted images (T1WI) (both axial and sagittal) and the maximum slice of the axial tumor lesion, respectively. Then, the synthetic minority oversampling technique (SMOTE) was utilized to balance the sample numbers. The optimal discriminative features were selected for model building. LightGBM algorithm was used to develop DLRM by a combination of radiomics features and deep learning features. For comparison, a radiomics model (RM) and a deep learning model (DLM) were constructed using a similar method as well. Differentiating efficacy was determined by using the receiver operating characteristic (ROC) analysis.
A total of 15 features were selected to construct the DLRM with SMOTE, which showed good discrimination performance in both the training and test cohorts. The DLRM outperformed RM and DLM for differentiating low- and high-grade meningiomas (training AUC: 0.988 vs. 0.980 vs. 0.892; test AUC: 0.935 vs. 0.918 vs. 0.718). The accuracy, sensitivity, and specificity of the DLRM with SMOTE were 0.926, 0.900, and 0.924 in the test cohort, respectively.
The DLRM with SMOTE based on enhanced T1WI images has favorable performance for noninvasively individualized prediction of meningioma grades, which exhibited favorable clinical usefulness superior over the radiomics features.
本研究旨在探讨基于增强 T1WI 的深度学习放射组学模型（DLRM）在区分低级别和高级别脑膜瘤中的临床应用价值。
共连续纳入 132 例经病理证实的脑膜瘤患者（训练队列 105 例，测试队列 27 例）。分别从 T1 加权图像（T1WI）（轴位和矢状位）和轴向肿瘤病变的最大切片中提取放射组学特征和深度学习特征。然后，利用合成少数过采样技术（SMOTE）平衡样本数量。选择最佳鉴别特征构建模型。使用 LightGBM 算法，通过放射组学特征和深度学习特征的组合，建立 DLRM。为了比较，还使用类似的方法建立了放射组学模型（RM）和深度学习模型（DLM）。通过受试者工作特征（ROC）分析确定区分效能。
利用 SMOTE 共选择了 15 个特征构建 DLRM，在训练和测试队列中均具有良好的鉴别性能。与 RM 和 DLM 相比，DLRM 可更好地区分低级别和高级别脑膜瘤（训练 AUC：0.988 比 0.980 比 0.892；测试 AUC：0.935 比 0.918 比 0.718）。在测试队列中，SMOTE 联合 DLRM 的准确率、敏感度和特异度分别为 0.926、0.900 和 0.924。
基于增强 T1WI 图像的 DLRM 具有良好的脑膜瘤分级无创个体化预测性能，临床应用价值优于放射组学特征。
