Abstract
OBJECTIVES: EGFR testing is a mandatory step before targeted therapy for non-small cell lung cancer patients. Combining some quantifiable features to establish a predictive model of EGFR expression status, break the limitations of tissue biopsy. MATERIALS AND METHODS: We retrospectively analyzed 1074 patients of non-small cell lung cancer with complete reports of EGFR gene testing. Then manually segmented VOI, captured the clinicopathological features, analyzed traditional radiology features, and extracted radiomic, and deep learning features. The cases were randomly divided into training and test set. We carried out feature screening; then applied the light GBM algorithm, Resnet-101 algorithm, logistic regression to develop sole models, and fused models to predict EGFR mutation conditions. The efficiency of models was evaluated by ROC and PRC curves. RESULTS: We successfully established Model(clinical), Model(radiomic), Model(CNN) (based on clinical-radiology, radiomic and deep learning features respectively), Model(radiomic+clinical) (combining clinical-radiology and radiomic features), and Model(CNN+radiomic+clinical) (combining clinical-radiology, radiomic, and deep learning features). Among the prediction models, Model(CNN+radiomic+clinical) showed the highest performance, followed by Model(CNN), and then Model(radiomic+clinical). All three models were able to accurately predict EGFR mutation with AUC values of 0.751, 0.738, and 0.684, respectively. There was no significant difference in the AUC values between Model(CNN+radiomic+clinical) and Model(CNN). Further analysis showed that Model(CNN+radiomic+clinical) effectively improved the efficacy of Model(radiomic+clinical) and showed better efficacy than Model(CNN). The inclusion of clinical-radiology features did not effectively improve the efficacy of Model(radiomic). CONCLUSIONS: Either deep learning or radiomic signature-based models can provide a fairly accurate non-invasive prediction of EGFR expression status. The model combined both features effectively enhanced the performance of radiomic models and provided marginal enhancement to deep learning models. Collectively, fusion models offer a novel and more reliable way of providing the efficacy of currently developed prediction models, and have far-reaching potential for the optimization of noninvasive EGFR mutation status prediction methods.