Department of Physiology, Pharmacology and Medical Physics, Alborz University of Medical Sciences, Karaj, Iran.
Department of Biomedical Engineering and Medical Physics, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Endocrine. 2023 Nov;82(2):326-334. doi: 10.1007/s12020-023-03407-6. Epub 2023 Jun 9.
This study aims to use ultrasound derived features as biomarkers to assess the malignancy of thyroid nodules in patients who were candidates for FNA according to the ACR TI-RADS guidelines.
Two hundred and ten patients who met the selection criteria were enrolled in the study and subjected to ultrasound-guided FNA of thyroid nodules. Different radiomics features were extracted from sonographic images, including intensity, shape, and texture feature sets. Least Absolute Shrinkage and Selection Operator (LASSO), Minimum Redundancy Maximum Relevance (MRMR), and Random Forests/Extreme Gradient Boosting Machine (XGBoost) algorithms were used for feature selection and classification of the univariate and multivariate modeling, respectively. Evaluation of models performed using accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).
In the univariate analysis, Gray Level Run Length Matrix - Run-Length Non-Uniformity (GLRLM-RLNU) and gray-level zone length matrix - Run-Length Non-Uniformity (GLZLM-GLNU) (both with an AUC of 0.67) were top-performing for predicting nodules malignancy. In the multivariate analysis of the training dataset, the AUC of all combinations of feature selection algorithms and classifiers was 0.99, and the highest sensitivity was for XGBoost classifier and MRMR feature selection algorithms (0.99). Finally, the test dataset was used to evaluate our model in which XGBoost classifier with MRMR and LASSO feature selection algorithms had the highest performance (AUC = 0.95).
Ultrasound-extracted features can be used as non-invasive biomarkers for thyroid nodules' malignancy prediction.
本研究旨在利用超声衍生特征作为生物标志物，根据 ACR TI-RADS 指南评估行细针抽吸活检术（FNA）的甲状腺结节患者的恶性程度。
本研究纳入了 210 名符合选择标准的患者，并对其进行了超声引导下甲状腺结节 FNA。从超声图像中提取了不同的放射组学特征，包括强度、形状和纹理特征集。使用最小绝对收缩和选择算子（LASSO）、最小冗余最大相关性（MRMR）和随机森林/极端梯度提升机（XGBoost）算法分别进行特征选择和单变量及多变量建模的分类。采用准确性、敏感度、特异性和受试者工作特征曲线（ROC）下面积（AUC）评估模型。
在单变量分析中，灰度游程长度矩阵-游程长度非均匀性（GLRLM-RLNU）和灰度区域长度矩阵-游程长度非均匀性（GLZLM-GLNU）（AUC 均为 0.67）在预测结节恶性程度方面表现最佳。在训练数据集的多变量分析中，所有特征选择算法和分类器组合的 AUC 均为 0.99，XGBoost 分类器和 MRMR 特征选择算法的最高敏感度为 0.99。最后，使用测试数据集评估了我们的模型，其中 XGBoost 分类器与 MRMR 和 LASSO 特征选择算法的性能最高（AUC=0.95）。
超声提取的特征可作为甲状腺结节恶性程度预测的无创性生物标志物。
