Department of Radiology, Xiangya Hospital, Central South University, Changsha 410008.
School of Computer Science and Engineering, South China University of Technology, Guangzhou 510640, China.
Zhong Nan Da Xue Xue Bao Yi Xue Ban. 2021 Apr 28;46(4):385-392. doi: 10.11817/j.issn.1672-7347.2021.200074.
Glioma is the most common intracranial primary tumor in central nervous system. Glioma grading possesses important guiding significance for the selection of clinical treatment and follow-up plan, and the assessment of prognosis. This study aims to explore the feasibility of logistic regression model based on radiomics to predict glioma grading.
Retrospective analysis was performed on 146 glioma patients with confirmed pathological diagnosis from January, 2012 to December, 2018. A total of 41 radiomics features were extracted from contrast-enhanced T-weighted imaging (TWI+C) lesion by manual segmentation. Least absolute shrinkage and selection operator (LASSO) was used to select the most-predictive radiomics features for pathological grading and to calculate radiomics score (Rad-score) of each patient. A logistic regression model was built to explore the correlation between giloma grading and Rad-score. Receiver operating characteristic (ROC) curve was performed to evaluate the model's predictive ability with area under the curve (AUC) for the evaluation index. Hosmer-Lemeshow test was used to measure the model's predictive accuracy.
A total of 5 imaging features selected by LASSO were used to establish a logistic regression model for predicting glioma grading. The model showed good discrimination with AUC value of 0.919. Hosmer-Lemeshow test showed no significant difference between the calibration curve and the ideal curve (=0.808), indicating high predictive accuracy of the model.
The logistic regression model using radiomics exhibits a relatively high accuracy for predicting glioma grading, which may serve as a complementary tool for preoperative prediction of giloma grading.
脑胶质瘤是中枢神经系统最常见的颅内原发性肿瘤。胶质瘤分级对临床治疗方案的选择和随访计划以及预后评估具有重要的指导意义。本研究旨在探讨基于放射组学的逻辑回归模型预测脑胶质瘤分级的可行性。
回顾性分析 2012 年 1 月至 2018 年 12 月期间经病理证实的 146 例脑胶质瘤患者。通过手动分割对增强 T 加权成像（TWI+C）病灶进行 41 个放射组学特征提取。最小绝对收缩和选择算子（LASSO）用于选择对病理分级最具预测性的放射组学特征，并计算每位患者的放射组学评分（Rad-score）。建立逻辑回归模型，探讨胶质瘤分级与 Rad-score 之间的相关性。采用受试者工作特征（ROC）曲线评估模型的预测能力，以曲线下面积（AUC）作为评价指标。Hosmer-Lemeshow 检验用于测量模型的预测准确性。
LASSO 选择的 5 个影像学特征用于建立预测脑胶质瘤分级的逻辑回归模型。该模型具有良好的判别能力，AUC 值为 0.919。Hosmer-Lemeshow 检验表明，校准曲线与理想曲线之间无显著差异（=0.808），表明该模型具有较高的预测准确性。
使用放射组学的逻辑回归模型预测脑胶质瘤分级具有较高的准确性，可能成为术前预测脑胶质瘤分级的辅助工具。
