AUTHOR=Zhang Lu , Ge Yinghui , Gao Qiuru , Zhao Fei , Cheng Tianming , Li Hailiang , Xia Yuwei 

TITLE=Machine Learning-Based Radiomics Nomogram With Dynamic Contrast-Enhanced MRI of the Osteosarcoma for Evaluation of Efficacy of Neoadjuvant Chemotherapy

JOURNAL=Frontiers in Oncology

VOLUME=Volume 11 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.758921

DOI=10.3389/fonc.2021.758921

ISSN=2234-943X

ABSTRACT=Objectives: To evaluate the value of machine learning-based Dynamic Contrast-Enhanced MRI(DCE-MRI) radiomics nomogram in prediction treatment response of Neoadjuvant Chemotherapy (NAC) in patients with osteosarcoma.
Methods: 102 patients with osteosarcoma underwent NAC were enrolled in this study. All patients received a DCE-MRI scan before NAC. The Response Evaluation Criteria in Solid Tumors (RECIST) was used as the standard to evaluate the NAC response in effective  and ineffective group. The following semi-quantitative parameters of DCE-MRI were calculated: early dynamic enhancement wash-in slope (Slope), time to peak (TTP), and enhancement rate (R). Variance Threshold, Univariate feature selection, and Least Absolute Shrinkage and Selection Operator (LASSO) were used to select optimal features. Three classifiers (K-Nearest Neighbor (KNN), Support Vector Machin (SVM), and Logistic Regression (LR)) were implemented for model establishment. The performance of different classifiers and semi-quantitative parameters was evaluated by confusion matrix and Receiver Operating Characteristics (ROC) curves. Furthermore, clinically relevant risk factors including age, tumor size and site, pathological fracture, and surgical staging were collected. The selected clinical features and imaging features were combined to establish the model and the Nomogram, and then the predictive efficacy was evaluated.
Results: Clinical relevance risk factor analysis demonstrates that only surgical stage was an independent predictor of NAC. A total of 7 radiomic features were selected, and three machine learning models (KNN, SVM, and LR) were established based on such features. Prediction accuracy (ACC) of these 3 models was 0.89, 0.84, and 0.84, respectively. And the area under the subject curve (AUC) of these 3 models was 0.86, 0.92, and 0.93 respectively. As for Slope, TTP, and R parameters, the prediction ACC was 0.91, 0.89, and 0.81, respectively; while the AUC was 0.87, 0.85, and 0.83, respectively. In both the training and testing sets, the ACC and AUC of the combined model were higher than those of the radiomics models (ACC = 0.91 and AUC = 0.95), which indicates outstanding performance of  proposed model. 
Conclusions: The radiomics nomogram demonstrates satisfactory predictive results for treatment response of patients with osteosarcoma before NAC. This finding may provide a new decision basis to improve the treatment plan.