Differential Diagnostic Performance Of The Predictive Model Combining BI-RADS Classification And Radiomic Classifier For Various X-ray Phenotype Of Breast Lesions

Objective: To evaluate the differential diagnostic performance of a predictive model combining BI-RADS classification and mammography radiomic classifier for various X-ray phenotype of breast lesions(mass,calcification,asymmetry,and structural distortion).Material and Methods: A total of 2 055 female patients were examined with mammography and the pathological diagnosis results were obtained from May 2013 to August 2020 at Zhong Da Hospital of Southeast University.Radiologists classified each lesion into mass and non-mass according to the fifth edition of BI-RADS.The mass was further divided into small mass(maximum diameter ≤ 2 cm)and large mass(maximum diameter > 2 cm),the non-mass was further divided into asymmetric,calcification and structural distortion.The radiomics model was constructed by the radiomics features being extracted in the region of interest of each lesion which segmented manually.Receiver operating characteristic(ROC)curve was used to assess the diagnostic efficacy of the BI-RADS classification,the radiomics model and the combined model for various phenotypes of breast lesions.Differences in the area under the ROC curve(AUC)were analyzed by the De Long test.Result:1.The combined model showed the best performance to differentiate benign and malignant lesions compared to BI-RADS category and the radiomics(AUC =0.947±0.005,0.924±0.006 and 0.827±0.009),the difference was statistically significant(Z = 9.29,14.94,P< 0.001).2.For large mass,small mas,non-mass,combined model(AUC = 0.958±0.007,0.933±0.013,0.939±0.008)showed the best performance when compared to the BI-RADS classification(AUC = 0.937±0.010,0.896±0.020,0.916±0.011;Z=5.32,3.90,5.08,P<0.001)or the radiomics model(AUC = 0.872±0.012,0.851±0.021,0.758±0.016;Z=7.86,4.53,12.13,P<0.001).3.The AUC of the combined model in diagnosing benign and malignant asymmetric breast lesions(0.897±0.017)was higher than that of the BI-RADS classification(AUC = 0.866±0.020,Z = 4.27,P < 0.001)and the radiomics model(AUC =0.633±0.029,Z = 7.44,P < 0.001).4.However,the AUC of the combined model in diagnosing benign and malignant calcified and structural distortions breast lesions(0.971±0.010,0.811±0.057,respectively)was only higher than that of the radiomics model(AUC = 0.827±0.021,0.586±0.075,Z = 7.40,3.15,P < 0.001),and there was no significant difference with the BI-RADS classification(AUC = 0.959±0.012,0.800±0.061,Z=1.87,0.39,P > 0.05).Conclusion:The combined model showed better differential diagnostic performance than simple models,and demonstrated good clinical usefulness.