基于磁共振的放射组学在鉴别成人眼内肿块中葡萄膜黑色素瘤与其他病变的价值。
Value of MR-based radiomics in differentiating uveal melanoma from other intraocular masses in adults.
机构信息
Department of Radiology, Beijing Tongren Hospital, Capital Medical University, No. 1 Dongjiaominxiang Street, Dongcheng District, Beijing, 100730, China; Clinical Center for Eye Tumors, Capital Medical University, Beijing, 100730, China.
Clinical Center for Eye Tumors, Capital Medical University, Beijing, 100730, China; Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, No. 1 Dongjiaominxiang Stree, Dongcheng District, Beijing, 100730, China; Beijing Key Laboratory of Intraocular Tumor Diagnosis and Treatment, Beijing 100730, China.
出版信息
Eur J Radiol. 2020 Oct;131:109268. doi: 10.1016/j.ejrad.2020.109268. Epub 2020 Sep 8.
PURPOSE
To assess the performance of machine learning (ML)-based magnetic resonance imaging (MRI) radiomics analysis for discriminating between uveal melanoma (UM) and other intraocular masses.
METHODS
This retrospective study analyzed 245 patients with intraocular masses (165 UMs and 80 other intraocular masses). Radiomics features were extracted from T1WI, T2WI, and contrast enhanced T1-weighted images (CET1WI), respectively. The intraclass correlation coefficient (ICC) was calculated to quantify the reproducibility of features. The training and test sets consisted of 195 and 50 cases. Least absolute shrinkage and selection operator (LASSO) regression method was employed for feature selection. The ML classifiers were logistic regression (LR), multilayer perceptron (MLP), and support vector machine (SVM). The performance of classifiers was evaluated by ROC analysis, and was compared to the performance of visual assessment by DeLong test.
RESULTS
The optimal radiomics feature set was 10, 15, 15, and 24 for T1W, T2W, CET1W, and joint T2W and CET1W images, respectively. The accuracy of T1WI, T2WI, CET1WI, and the joint T2WI and CET1WI models ranged from 72.0 %-78.0 %, from 79.6 %-81.6 %, from 74.0 %-82.0 %, and from 76.0 %-86.0 % in the test set. In the test set, the AUC for T1WI, T2WI, CET1WI, joint T2WI, and CET1WI models ranged from 0.775 to 0.829, 0.816 to 0.826, 0.836 to 0.861, and 0.870 to 0.877, respectively. In the combined model, the performance of ML classifiers was better than the performance of visual assessment in the training set and in all patients (p＜0.05).
CONCLUSIONS
Radiomics analysis represents a promising tool in separating UM from other intraocular masses.
目的
评估基于机器学习（ML）的磁共振成像（MRI）放射组学分析在鉴别葡萄膜黑色素瘤（UM）和其他眼内肿块中的性能。
方法
本回顾性研究分析了 245 名眼内肿块患者（165 例 UM 和 80 例其他眼内肿块）。分别从 T1WI、T2WI 和对比增强 T1 加权图像（CET1WI）中提取放射组学特征。采用组内相关系数（ICC）量化特征的可重复性。训练集和测试集分别包含 195 例和 50 例病例。采用最小绝对收缩和选择算子（LASSO）回归方法进行特征选择。ML 分类器为逻辑回归（LR）、多层感知机（MLP）和支持向量机（SVM）。通过 ROC 分析评估分类器的性能，并通过 DeLong 检验与视觉评估的性能进行比较。
结果
T1W、T2W、CET1W 和 T2W 与 CET1W 联合图像的最佳放射组学特征集分别为 10、15、15 和 24。在测试集中，T1WI、T2WI、CET1WI 和 T2W 与 CET1WI 模型的准确率范围分别为 72.0%-78.0%、79.6%-81.6%、74.0%-82.0%和 76.0%-86.0%。在测试集中，T1WI、T2WI、CET1WI、T2W 与 CET1WI 模型的 AUC 范围分别为 0.775-0.829、0.816-0.826、0.836-0.861 和 0.870-0.877。在联合模型中，ML 分类器在训练集和所有患者中的性能均优于视觉评估（p＜0.05）。
结论
放射组学分析是一种有前途的工具，可用于将 UM 与其他眼内肿块区分开来。
Zotero 插件