The Value Of CEM Intratumoral And Peritumoral Imaging In The Diagnosis Of Breast Cancer And Prediction Of HER-2 Statu

Chapter one: Intra-and peritumoral radiomics for predicting malignant BI-RADS category four breast lesions on contrast-enhanced spectral mammography: a multicenter studyObjective: To construct and test a nomogram based on intra-and peritumoral radiomics and clinical factors for predicting malignant BI-RADS four lesions on contrast-enhanced mammography(CEM).Methods: A total of 884 patients with BI-RADS four lesions were collected from two centers.The radiomics features were extracted from the intratumoral region(ITR),peritumoral regions(PTRs)of 5mm and 10 mm around the tumor,and ITR plus PTRs of 5 mm and 10 mm and five radiomics signatures were established after selecting features by minimum-redundancy-maximum-relevance and least absolute shrinkage and selection operator regression.A nomogram was built using selected signatures and clinical factors by logistic regression analysis.The performance of the nomogram was assessed with the ROC,decision curve analysis,and calibration curves,and also compared with the radiomics model,clinical model,and radiologists.Results: The nomogram built by three radiomics signatures(constructed from ITR,5 mm PTR,and ITR + 10 mm PTR)and two clinical factors(age and BI-RADS category)showed powerful predictive ability in internal and external test sets with AUCs of 0.907 and 0.904,respectively,which were higher than 0.826(P<0.01)and0.827(P=0.15)of radiomics model and 0.885(P=0.29)and 0.847(P=0.22)of clinical model.The calibration curves,DCA showed favorable predictive performance of the nomogram.In addition,radiologists improved diagnostic performance with the help of nomogram.Conclusion: Nomogram based on intratumoral and peritumoral radiomics features combined with clinical risk variables has a certain value in identifying benign and malignant BI-RADS four lesions.Chapter two: Intra-and peritumoral radiomics for predicting HER-2 IHC2+ status in breast cancer on contrast-enhanced mammographyObjective: To construct and test a nomogram based on intra-and peritumoral radiomics features and clinical factors for predicting the equivocal HER-2(IHC 2+)status of breast cancer patients using CEM.Methods: A total of 106 female patients with IHC 2+ of breast cancer were retrospectively enrolled in the study and grouped into training(n=84)and internal test sets(n=22).In addition,26 patients consist of a prospective test cohort.Radiomics features were extracted from intratumoral and peritumoral regions on CEM and were selected by low variance and least absolute shrinkage and selection operator regression(LASSO).Five radiomcis signatures were established from the intratumoral region,peritumoral regions of 5 and 10 mm around the tumor,and intratumoral plus peritumoral regions of 5 and 10 mm using LASSO regression.A nomogram was built using selected signatures and clinical factors by logistic regression analysis.The prediction performance of nomogram was assessed by receiver operator characteristic curve,the calibration curve,and decision curve analysis and was compared with radiomics model and clinical model.Results: The nomogram included the intratumoral signature,5mm-peritumoral signature,and tumor diameter.In the internal test cohort,the AUC of nomogram was 0.893(95CI%:0.756-1.000),higher than the radiomics model with 0.821(95CI%:0.641-1.000)(P=0.292)and clinical model with 0.866(95CI%:0.681-1.000)(P=0.314).In the prospective test cohort,the AUC of nomogram was 0.840(95CI%:0.652-1.000),higher than the radiomics model with 0.819(95CI%:0.649-0.999)(P=0.676)and clinical model with 0.774(95CI%: 0.547-1.000)(P=0.878).The calibration curves,DCA showed favorable predictive performance of the nomogram.Conclusion: The nomogram incorporated the intratumoral and peritumoral radiomics signatures and clinical risk variables could be conveniently used to facilitate the preoperative individualized prediction of HER-2 IHC2+ status of breast cancer patients on CEM.