Yin Ping, Mao Ning, Chen Hao, Sun Chao, Wang Sicong, Liu Xia, Hong Nan
Department of Radiology, Peking University People's Hospital, Beijing, Beijing Municipality, China.
Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University, Yantai, China.
Front Oncol. 2020 Oct 16;10:564725. doi: 10.3389/fonc.2020.564725. eCollection 2020.
To assess the performance of deep neural network (DNN) and machine learning based radiomics on 3D computed tomography (CT) and clinical characteristics to predict benign or malignant sacral tumors.
This single-center retrospective analysis included 459 patients with pathologically proven sacral tumors. After semi-automatic segmentation, 1,316 hand-crafted radiomics features of each patient were extracted. All models were built on training set (321 patients) and tested on validation set (138 patients). A DNN model and four machine learning classifiers (logistic regression [LR], random forest [RF], support vector machine [SVM] and k-nearest neighbor [KNN]) based on CT features and clinical characteristics were built, respectively. The area under the receiver operating characteristic curve (AUC) and accuracy (ACC) were used to evaluate different models.
In total, 459 patients (255 males, 204 females; mean age of 42.1 ± 17.8 years, range 4-82 years) were enrolled in this study, including 206 cases of benign tumor and 253 cases of malignant tumor. The sex, age and tumor size had significant differences between the benign tumors and malignant tumors ( = 10.854, = -6.616, = 2.843, < 0.05). The radscore, sex, and age were important indicators for differentiating benign and malignant sacral tumors (odds ratio [OR]1 = 2.492, OR2 = 2.236, OR3 = 1.037, < 0.01). Among the four clinical-radiomics models (RMs), clinical-LR had the best performance in the validation set (AUC = 0.84, ACC = 0.81). The clinical-DNN model also achieved a high performance (an AUC of 0.83 and an ACC of 0.76 in the validation set) in identifying benign and malignant sacral tumors.
Both the clinical-LR and clinical-DNN models would have a high impact on assisting radiologists in their clinical diagnosis of sacral tumors.
评估基于深度神经网络（DNN）和机器学习的影像组学在三维计算机断层扫描（CT）及临床特征方面预测骶骨肿瘤良恶性的性能。
这项单中心回顾性分析纳入了459例经病理证实的骶骨肿瘤患者。在进行半自动分割后，提取了每位患者的1316个手工制作的影像组学特征。所有模型均基于训练集（321例患者）构建，并在验证集（138例患者）上进行测试。分别构建了基于CT特征和临床特征的DNN模型以及四个机器学习分类器（逻辑回归[LR]、随机森林[RF]、支持向量机[SVM]和k近邻[KNN]）。采用受试者操作特征曲线下面积（AUC）和准确率（ACC）来评估不同模型。
本研究共纳入459例患者（男性255例，女性204例；平均年龄42.1±17.8岁，范围4 - 82岁），其中良性肿瘤206例，恶性肿瘤253例。良性肿瘤与恶性肿瘤在性别、年龄和肿瘤大小方面存在显著差异（ = 10.854， = -6.616， = 2.843， < 0.05）。影像组学评分、性别和年龄是区分骶骨肿瘤良恶性的重要指标（优势比[OR]1 = 2.492，OR2 = 2.236，OR3 = 1.037， < 0.01）。在四个临床影像组学模型（RM）中，临床-LR在验证集中表现最佳（AUC = 0.84，ACC = 0.81）。临床-DNN模型在识别骶骨肿瘤良恶性方面也具有较高性能（验证集中AUC为0.83，ACC为0.76）。
临床-LR和临床-DNN模型在辅助放射科医生对骶骨肿瘤进行临床诊断方面均具有重要影响。
