Department of Radiology, Peking University People's Hospital, 11 Xizhimen Nandajie, Xicheng District, Beijing, 100044, People's Republic of China.
Department of Radiology, Qindao University Medical College Affiliated Yantai Yuhuangding Hospital, Yantai, Shandong, People's Republic of China.
Eur Radiol. 2019 Apr;29(4):1841-1847. doi: 10.1007/s00330-018-5730-6. Epub 2018 Oct 2.
We aimed to identify optimal machine-learning methods for preoperative differentiation of sacral chordoma (SC) and sacral giant cell tumour (SGCT) based on 3D non-enhanced computed tomography (CT) and CT-enhanced (CTE) features.
A total of 95 patients were divided into a training set and a validation set. Three best feature selection methods (Relief, least absolute shrinkage and selection operator (LASSO) and Random Forest (RF)) and three classification methods, including generalised linear models (GLM), support vector machines (SVM) and RF, were compared for their performance in distinguishing SC and SGCT. The performance of the radiomics model was investigated via area under the receiver-operating characteristic curve (AUC) and accuracy (ACC) analysis.
The selection method LASSO + classifier GLM had the highest AUC of 0.984 and ACC of 0.897 in the validating set, followed by Relief + GLM (AUC = 0.909, ACC = 0.862) and LASSO + SVM (AUC = 0.900, ACC = 0.862) based on CTE features. For CT features, RF + GLM had the highest AUC of 0.889, while LASSO + GLM achieved a high ACC of 0.793 in the validating set. Regardless of the methods, CTE features significantly outperformed those from CT for the differentiation of SC and SGCT (Z = -3.029, Z = -4.553; p < 0.05).
Our study demonstrated CTE features performed better than CT features. The selection method LASSO + classifier GLM had the best performance in differentiation of SC and SGCT, which could enhance the application of radiomics methods in sacral tumours.
• Sacral chordoma and sacral giant cell tumour are the two most common primary tumours of the sacrum with many common clinical and imaging characteristics. • A radiomics model helps clinicians to identify the histology of a sacral tumour. • CTE features should be preferred.
我们旨在基于三维非增强计算机断层扫描（CT）和 CT 增强（CTE）特征，确定用于术前区分骶骨脊索瘤（SC）和骶骨巨细胞瘤（SGCT）的最佳机器学习方法。
共 95 例患者被分为训练集和验证集。三种最佳特征选择方法（Relief、最小绝对值收缩和选择算子（LASSO）和随机森林（RF））和三种分类方法，包括广义线性模型（GLM）、支持向量机（SVM）和 RF，用于区分 SC 和 SGCT 的性能进行了比较。通过接收者操作特征曲线（AUC）和准确度（ACC）分析评估了放射组学模型的性能。
在验证集中，基于 CTE 特征，选择方法 LASSO+分类器 GLM 的 AUC 为 0.984，ACC 为 0.897，其次是 Relief+GLM（AUC=0.909，ACC=0.862）和 LASSO+SVM（AUC=0.900，ACC=0.862）。对于 CT 特征，RF+GLM 的 AUC 最高，为 0.889，而 LASSO+GLM 在验证集中的 ACC 为 0.793。无论采用何种方法，CTE 特征在区分 SC 和 SGCT 方面均明显优于 CT 特征（Z=-3.029，Z=-4.553；p<0.05）。
本研究表明 CTE 特征优于 CT 特征。选择方法 LASSO+分类器 GLM 在区分 SC 和 SGCT 方面具有最佳性能，这可以增强放射组学方法在骶骨肿瘤中的应用。
骶骨脊索瘤和骶骨巨细胞瘤是骶骨最常见的两种原发性肿瘤，具有许多共同的临床和影像学特征。
放射组学模型有助于临床医生识别骶骨肿瘤的组织学特征。
CTE 特征应该是首选。
