Hai Liao1, Xiaobo Chen2,3, Shaolu Lu4, Guanqiao Jin1, Wei Pei1, Ye Li5, Yunyun Wei1, Xia Huang4, Chenghuan Wang4, Xueli Liang1, Huayan Bao1, Lidong Liu1, Danke Su1. 1. Department of Medical Imaging Center, Guangxi Medical University Cancer Hospital, Nanning, Guangxi, China. 2. Department of Radiology, Guangdong Provincial People's Hospital/Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China. 3. Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China. 4. Department of Radiology, Wuzhou Red Cross Hospital, Wuzhou, Guangxi, China. 5. Department of Radiotherapy, Guangxi Medical University Cancer Hospital, Nanning, Guangxi, China.
Abstract
BACKGROUND: Pretreatment individualized assessment of tumor response to induction chemotherapy (ICT) is a need in locoregionally advanced nasopharyngeal carcinoma (LANPC). Imaging method plays vital role in tumor response assessment. However, powerful imaging method for ICT response prediction in LANPC is insufficient. PURPOSE: To establish a robust model for predicting response to ICT in LANPC by comparing the performance of back propagation neural network (BPNN) model with logistic regression model. STUDY TYPE: Retrospective. POPULATION: A total of 286 LANPC patients were assigned to training (N = 200, 43.8 ± 10.9 years, 152 male) and testing (N = 86, 43.5 ± 11.3 years, 57 male) cohorts. FIELD STRENGTH/SEQUENCE: T2 -weighted imaging, contrast enhanced-T1 -weighted imaging using fast spin echo sequences at 1.5 T scanner. ASSESSMENT: Predictive clinical factors were selected by univariate and multivariate logistic models. Radiomic features were screened by interclass correlation coefficient, single-factor analysis, and the least absolute shrinkage selection operator (LASSO). Four models based on clinical factors (Modelclinic ), radiomics features (Modelradiomics ), and clinical factors + radiomics signatures using logistic (Modelcombined ), and BPNN (ModelBPNN ) methods were established, and model performances were compared. STATISTICAL TESTS: Student's t-test, Mann-Whitney U-test, and Chi-square test or Fisher's exact test were used for comparison analysis. The performance of models was assessed by area under the receiver operating characteristic (ROC) curve (AUC) and Delong test. P < 0.05 was considered statistical significance. RESULTS: Three significant clinical factors: Epstein-Barr virus-DNA (odds ratio [OR] = 1.748; 95% confidence interval [CI], 0.969-3.171), sex (OR = 2.883; 95% CI, 1.364-6.745), and T stage (OR = 1.853; 95% CI, 1.201-3.052) were identified via univariate and multivariate logistic models. Twenty-four radiomics features were associated with treatment response. ModelBPNN demonstrated the highest performance among Modelcombined , Modelradiomics , and Modelclinic (AUC of training cohort: 0.917 vs. 0.808 vs. 0.795 vs. 0.707; testing cohort: 0.897 vs. 0.755 vs. 0.698 vs. 0.695). CONCLUSION: A machine-learning approach using BPNN showed better ability than logistic regression model to predict tumor response to ICT in LANPC. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.
BACKGROUND: Pretreatment individualized assessment of tumor response to induction chemotherapy (ICT) is a need in locoregionally advanced nasopharyngeal carcinoma (LANPC). Imaging method plays vital role in tumor response assessment. However, powerful imaging method for ICT response prediction in LANPC is insufficient. PURPOSE: To establish a robust model for predicting response to ICT in LANPC by comparing the performance of back propagation neural network (BPNN) model with logistic regression model. STUDY TYPE: Retrospective. POPULATION: A total of 286 LANPC patients were assigned to training (N = 200, 43.8 ± 10.9 years, 152 male) and testing (N = 86, 43.5 ± 11.3 years, 57 male) cohorts. FIELD STRENGTH/SEQUENCE: T2 -weighted imaging, contrast enhanced-T1 -weighted imaging using fast spin echo sequences at 1.5 T scanner. ASSESSMENT: Predictive clinical factors were selected by univariate and multivariate logistic models. Radiomic features were screened by interclass correlation coefficient, single-factor analysis, and the least absolute shrinkage selection operator (LASSO). Four models based on clinical factors (Modelclinic ), radiomics features (Modelradiomics ), and clinical factors + radiomics signatures using logistic (Modelcombined ), and BPNN (ModelBPNN ) methods were established, and model performances were compared. STATISTICAL TESTS: Student's t-test, Mann-Whitney U-test, and Chi-square test or Fisher's exact test were used for comparison analysis. The performance of models was assessed by area under the receiver operating characteristic (ROC) curve (AUC) and Delong test. P < 0.05 was considered statistical significance. RESULTS: Three significant clinical factors: Epstein-Barr virus-DNA (odds ratio [OR] = 1.748; 95% confidence interval [CI], 0.969-3.171), sex (OR = 2.883; 95% CI, 1.364-6.745), and T stage (OR = 1.853; 95% CI, 1.201-3.052) were identified via univariate and multivariate logistic models. Twenty-four radiomics features were associated with treatment response. ModelBPNN demonstrated the highest performance among Modelcombined , Modelradiomics , and Modelclinic (AUC of training cohort: 0.917 vs. 0.808 vs. 0.795 vs. 0.707; testing cohort: 0.897 vs. 0.755 vs. 0.698 vs. 0.695). CONCLUSION: A machine-learning approach using BPNN showed better ability than logistic regression model to predict tumor response to ICT in LANPC. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.