Evaluation Of Mammography,Ultrasonography,MRI And Combined Inspection With ROC Curve In Differential Diagnosis Of Benign And Malignant Breast Masses

[Research purposes]1.To investigate the diagnostic value of mammography,ultrasonography and their combination in the differential diagnosis of benign and malignant breast masses.2.To investigate the diagnostic efficacies of 3.0T MRI DCE-lesion morphology,DCE-TIC curves,and DWI-ADC values in benign and malignant breast masses.3.ROC curve of multi-parametric magnetic resonance imaging and mammography,ultrasonography and their combination of diagnostic and comparative analysis of breast cancer benign and malignant tumors.Part 1ROC curve evaluation of mammography,ultrasound and their combined diagnosis of benign and malignant breast masses[Objective and Methods]1.Research object:Collect our hospital from June 2015 to October 2017 at the same time line mammography,ultrasonography and 3.0 T MRI and confirmed by surgery or biopsy pathology of 46 patients with breast lumps,a total of 49 lesions,including 25 malignant lesions,24 benign lesions.2.Equipment and inspection methods:2.1 Mammography equipment using GE 2000D models.2.2 Color Doppler ultrasound device using SAMSUNG RS80A and GE S8-type ultrasound system.3.Interpretation and statistical processing:The diagnosis is based on the classification of the North American Radiological Society(ACR)Breast Imaging Reporting and Data System(BI-RADS).The diagnostic imaging results remove the need for further examination of category 0,negative diagnosis of category I,and pathologically confirmed category VI of the breast mass.The diagnosis results were divided into 4 categories(?,?,?,and V).Types II to III were interpreted as benign lesions.Types IV to V were malignant lesions and compared with pathological criteria as a gold standard.Using SPSS 22.0 software,count data using x2 test(if not,using Fishers exact probability method)analysis and comparison of mammography,ultrasonography and their combined examination on the diagnosis of breast benign and malignant lesions,the difference was statistically significant;and draw three The ROC curve of the inspection method yields the area under the curve(AUC).[Results]1.The sensitivity of the mammography in the diagnosis of benign and malignant breast masses was 92.0%,the specificity was 79.2%,the accuracy was 85.7%,the positive predictive value was 82.1%,the negative predictive value was 90.5%,and the Kappa value was 0.713.2.The sensitivity of ultrasonography diagnosis of benign and malignant breast masses was 92.0%,specificity was 58.3%,accuracy was 75.5%,positive predictive value was 69.7%,negative predictive value was 87.5%and Kappa was 0.512.3.The sensitivity of mammography combined with ultrasonography in diagnosis of benign and malignant breast masses was 96.0%,specificity was 83.3%,accuracy was 89.8%,positive predictive value was 85.7%,negative predictive value was 95.2%,and Kappa value was 0.795.4.The study of this group showed that the AUC of the molybdenum target combined with ultrasound diagnosis was 0.897,the AUC of the molybdenum target diagnosis was 0.856,and the AUC of the ultrasound diagnosis was 0.752.[Conclusion]1.Breast mammography and ultrasonography are of great significance in the differential diagnosis of breast masses,and the consistency between the diagnosis and pathological findings of benign and malignant breast masses is higher.2.The results of ROC curve analysis showed that the combined diagnosis of mammography and ultrasonography can effectively make up for weaknesses and improve diagnostic efficiency;Compared with pathological results,the diagnosis of benign and malignant breast tumors High diagnostic consistency.part 2Multi-parameter magnetic resonance imaging for the differential diagnosis of benign and malignant breast masses and analysis of ROC curve compared with mammography,ultrasonography and their combined examination[Objective and Methods]1.Research Object:Same as the first part2.Equipment and Methods2.1 Mammography,color Doppler ultrasound equipment and diagnostic methods with the first part.2.2 MRI:This study used PHLIPS Achieva 3.0T dual-gradient superconducting MRI.2.2.1 Routine Scan Sequence:Including axial TSE T1WI sequence,transverse TSE T2WI SPAIR sequence.2.2.2 Dynamic Enhancement Sequence(DCE):Dyn-eTHRIVE SENSE axial scanning with T1 high resolution was used,and Gd-DTPA contrast agent was selected.Immediately after the completion of the scan T1WI sagittal position.2.2.3 Diffusion Weighted Imaging(DWI):The single-shot plane echo(SS-EPI)technique was used to examine the cross-sectional area.The b-value was 800 s/mm2 and the ADC value was measured.3.Interpretation and Statistical Processing:With the first part.Using SPSS 22.0 software,measurement data are expressed in the form of mean ±standard deviation(x±s),using independent sample t test;count data using)?2 test(if not,using Fishers exact probability method);draw MRI imaging diagnosis The ROC curve plot yields the area under the curve(AUC)and is compared with the molybdenum target,ultrasound and their combined examination.[Results]1.DCE-morphology was used as a criterion for benign and malignant masses.The positive predictive value for diagnosis was 82.8%,negative predictive value was 95.0%,sensitivity was 96.0%,specificity was 79.2%,accuracy was 87.8%,and Kappa was 0.754.2.The positive predictive value of the DCE-TIC curve as a criterion for benign and malignant masses was 84.2%,negative predictive value was 70.0%,sensitivity was 64.0%,specificity was 87.5%,accuracy was 75.5%,and Kappa was 0.512.3.The use of ROC curve analysis to determine the diagnosis of breast masses of benign and malignant ADC critical point of 0.962 × 10-3mm2/s,above the threshold is interpreted as benign,less than or equal to the threshold is interpreted as malignant,which is positive for the diagnosis of benign and malignant breast masses The predicted value was 88.0%,the negative predictive value was 87.5%,the sensitivity was 88.0%,the specificity was 87.5%,the accuracy was 87.8%,and the Kappa value was 0.755.4.Multi-parametric MRI combined examination was used to differentiate between benign and malignant breast lesions with a positive predictive value of 92.0%,a negative predictive value of 91.7%,a sensitivity of 92.0%,a specificity of 91.7%,an accuracy of 91.8%,and Kappa.The value was 0.837;The AUC for MRI diagnosis is 0.918.[Conclusion]1.3.0T magnetic resonance DCE-morphology,DCE-TIC curve and DWI-ADC values are of great significance in differential diagnosis of benign and malignant breast tumors.2.According to the pathological results,the DCE-morphology and DWI-ADC values are highly consistent with the diagnosis of benign and malignant breast masses;the consistency of the DCE-TIC curve is higher.3.The diagnostic efficiency of multi-parameter MRI imaging is higher than that of mammography,ultrasonography and their combination.