Department of Radiology, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, 639 Zhizaoju Road, Shanghai, 200011, China.
Eur Radiol. 2021 Sep;31(9):6429-6437. doi: 10.1007/s00330-021-07731-1. Epub 2021 Feb 10.
To develop and compare several machine learning models to predict occult cervical lymph node (LN) metastasis in early-stage oral tongue squamous cell cancer (OTSCC) from preoperative MRI texture features.
We retrospectively enrolled 116 patients with early-stage OTSCC (cT1-2N0) who had been surgically treated by tumor excision and elective neck dissection (END). For each patient, we extracted 86 texture features from T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (ceT1WI), respectively. Dimension reduction was performed in three consecutive steps: reproducibility analysis, collinearity analysis, and information gain algorithm. Models were created using six machine learning methods, including logistic regression (LR), random forest (RF), naïve Bayes (NB), support vector machine (SVM), AdaBoost, and neural network (NN). Their performance was assessed using tenfold cross-validation.
Occult LN metastasis was pathologically detected in 42.2% (49/116) of the patients. No significant association was identified between node status and patients' gender, age, or clinical T stage. Dimension reduction steps selected 6 texture features. The NB model gave the best overall performance, which correctly classified the nodal status in 74.1% (86/116) of the carcinomas, with an AUC of 0.802.
Machine learning-based MRI texture analysis offers a feasible tool for preoperative prediction of occult cervical node metastasis in early-stage OTSCC.
• A machine learning-based MRI texture analysis approach was adopted to predict occult cervical node metastasis in early-stage OTSCC with no evidence of node involvement on conventional images. • Six texture features from T2WI and ceT1WI of preoperative MRI were selected to construct the predictive model. • After comparing six machine learning methods, naïve Bayes (NB) achieved the best performance by correctly identifying the node status in 74.1% of the patients, using tenfold cross-validation.
从术前 MRI 纹理特征中开发并比较几种机器学习模型，以预测早期口腔舌鳞状细胞癌（OTSCC）隐匿性颈部淋巴结（LN）转移。
我们回顾性纳入了 116 例接受手术切除和选择性颈部清扫术（END）治疗的早期 OTSCC（cT1-2N0）患者。为每位患者分别从 T2 加权成像（T2WI）和对比增强 T1 加权成像（ceT1WI）中提取 86 个纹理特征。在三个连续步骤中进行降维：重复性分析、共线性分析和信息增益算法。使用六种机器学习方法（逻辑回归（LR）、随机森林（RF）、朴素贝叶斯（NB）、支持向量机（SVM）、AdaBoost 和神经网络（NN））建立模型。使用十折交叉验证评估模型性能。
42.2%（49/116）的患者病理检查发现隐匿性 LN 转移。淋巴结状态与患者性别、年龄或临床 T 分期之间无显著相关性。降维步骤选择了 6 个纹理特征。NB 模型的整体性能最佳，正确分类了 74.1%（86/116）的癌灶，AUC 为 0.802。
基于机器学习的 MRI 纹理分析为预测早期 OTSCC 隐匿性颈部淋巴结转移提供了一种可行的工具。
• 采用基于机器学习的 MRI 纹理分析方法，对术前无常规图像上无淋巴结受累证据的早期 OTSCC 进行隐匿性颈部淋巴结转移预测。• 从术前 MRI 的 T2WI 和 ceT1WI 中选择 6 个纹理特征构建预测模型。• 在比较了六种机器学习方法后，NB 通过十折交叉验证正确识别了 74.1%的患者的节点状态，性能最佳。
