Wei Qiurong, Chen Zeli, Tang Yehuan, Chen Weicui, Zhong Liming, Mao Liting, Hu Shaowei, Wu Yuankui, Deng Kan, Yang Wei, Liu Xian
Department of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, 510120, China.
Guangdong Provincial Key Laboratory of Medical Image Processing, School of Biomedical Engineering, Southern Medical University, Guangzhou, 510515, China.
Eur Radiol. 2023 Mar;33(3):1906-1917. doi: 10.1007/s00330-022-09204-5. Epub 2022 Nov 10.
The aim of this study was two-fold: (1) to develop and externally validate a multiparameter MR-based machine learning model to predict the pathological complete response (pCR) in locally advanced rectal cancer (LARC) patients after neoadjuvant chemoradiotherapy (nCRT), and (2) to compare different classifiers' discriminative performance for pCR prediction.
This retrospective study includes 151 LARC patients divided into internal (centre A, n = 100) and external validation set (centre B, n = 51). The clinical and MR radiomics features were derived to construct clinical, radiomics, and clinical-radiomics model. Random forest (RF), support vector machine (SVM), logistic regression (LR), K-nearest neighbor (KNN), naive Bayes (NB), and extreme gradient boosting (XGBoost) were used as classifiers. The predictive performance was assessed using the receiver operating characteristic (ROC) curve.
Eleven radiomics and four clinical features were chosen as pCR-related signatures. In the radiomics model, the RF algorithm achieved 74.0% accuracy (an AUC of 0.863) and 84.4% (an AUC of 0.829) in the internal and external validation sets. In the clinical-radiomics model, RF algorithm exhibited high and stable predictive performance in the internal and external validation datasets with an AUC of 0.906 (87.3% sensitivity, 73.7% specificity, 76.0% accuracy) and 0.872 (77.3% sensitivity, 88.2% specificity, 86.3% accuracy), respectively. RF showed a better predictive performance than the other classifiers in the external validation datasets of three models.
The multiparametric clinical-radiomics model combined with RF algorithm is optimal for predicting pCR in the internal and external sets, and might help improve clinical stratifying management of LARC patients.
• A two-centre study showed that radiomics analysis of pre- and post-nCRT multiparameter MR images could predict pCR in patients with LARC. • The combined model was superior to the clinical and radiomics model in predicting pCR in locally advanced rectal cancer. • The RF classifier performed best in the current study.
本研究有两个目的：（1）开发并外部验证基于多参数磁共振成像（MR）的机器学习模型，以预测局部晚期直肠癌（LARC）患者在新辅助放化疗（nCRT）后的病理完全缓解（pCR）；（2）比较不同分类器对pCR预测的判别性能。
这项回顾性研究纳入了151例LARC患者，分为内部验证集（中心A，n = 100）和外部验证集（中心B，n = 51）。提取临床和MR影像组学特征，构建临床、影像组学和临床-影像组学模型。使用随机森林（RF）、支持向量机（SVM）、逻辑回归（LR）、K近邻（KNN）、朴素贝叶斯（NB）和极端梯度提升（XGBoost）作为分类器。使用受试者工作特征（ROC）曲线评估预测性能。
选择了11个影像组学特征和4个临床特征作为与pCR相关的特征。在影像组学模型中，RF算法在内部和外部验证集中的准确率分别为74.0%（AUC为0.863）和84.4%（AUC为0.829）。在临床-影像组学模型中，RF算法在内部和外部验证数据集中表现出高且稳定的预测性能，AUC分别为0.906（灵敏度87.3%，特异性73.7%，准确率76.0%）和0.872（灵敏度77.3%，特异性88.2%，准确率86.3%）。在三个模型的外部验证数据集中，RF的预测性能优于其他分类器。
多参数临床-影像组学模型结合RF算法在内部和外部验证集中预测pCR效果最佳，可能有助于改善LARC患者的临床分层管理。
• 一项两中心研究表明，对nCRT前后的多参数MR图像进行影像组学分析可预测LARC患者的pCR。• 在预测局部晚期直肠癌的pCR方面，联合模型优于临床和影像组学模型。• 在本研究中，RF分类器表现最佳。
