Diagnostic Value Study Of Strain Elastography For BI-RADS Category 4 Lesions

Background Breast cancer has surpassed lung cancer as the most prevalent malignant tumor in women,and its incidence rate and fatality rate are rising annually in China.The quality of life and survival rate of individuals with breast cancer can be enhanced by early identification and accurate diagnosis.Breast BI-RADS category 4 lesions are suspected malignant nodules,and their malignant risk span is large(2%～95%),which brings great uncertainty to the clinician’s diagnosis and treatment plan.Evaluation of such nodules through elastic imaging technology can reduce unnecessary puncture biopsy of benign nodules and increase the confidence in the diagnosis of malignant nodules.Objective To investigate the diagnostic value of strain elastography in BI-RADS category 4 lesions.Methods A total of 224 patients(259 lesions in all)who were diagnosed as BI-RADS category 4 lesions by using Samsung RS80 A ultrasound diagnostic equipment in the People’s Hospital of China Three Gorges University · the First People’s Hospital of Yichang from July2021 to December 2022 were selected as the research objects.All patients with BI-RADS category 4 lesions were examined by elastic score method and strain ratio method,and the BI-RADS classification of lesions was modified according to elastic score method.Taking the pathological outcomes as the gold standard,draw the Receiver Operating Characteristic Curve(ROC)to obtain the best threshold for the diagnosis of benign and malignant breast lesions by the strain ratio method firstly.Secondly,the factors with statistically significant differences in clinical data,ultrasonic characteristics and elastic indexes in the benign and malignant groups were screened to construct a binary logistic regression model.Finally,calculate the sensitivity(SEN),specificity(SPE),positive predictive value(PPV),negative predictive value(NPV)and accuracy(ACC)of conventional ultrasound,elastic score approach,strain ratio technique,combined group and logistic regression model in the diagnosis of BI-RADS category 4lesions,draw the ROC of the five methods and compare the AUC under the curve to evaluate the diagnostic value,P<0.05 was statistical significant.Results(1)Among 259 BI-RADS category 4 lesions,163 were benign and 96 were malignant.Amongst them,159 benign and 100 malignant nodules were diagnosed by conventional ultrasound;185 benign and 74 malignant nodules were diagnosed by elastic score method;The strain ratio method used 2.30 as the best threshold to diagnose 162 benign and 97 malignant nodules;164 benign and 95 malignant nodules were diagnosed in the combined group;Logistic regression model diagnosed 168 benign and 91 malignant nodules.(2)The SEN,SPE,PPV,NPV,ACC and AUC of the BI-RADS category 4 lesions diagnosed by conventional ultrasound were 87.50%,90.18%,84.00%,92.45%,89.19% and 0.888 respectively.The SEN,SPE,PPV,NPV,ACC and AUC of the category 4 lesions diagnosed by elastic score method were 73.96%,98.16%,95.95%,86.49%,89.19% and 0.861 respectively.The SEN,SPE,PPV,NPV,ACC and AUC of the category 4 lesions diagnosed by strain ratio method was 89.58%,93.25%,88.66%,93.83%,91.89% and 0.914 respectively.The SEN,SPE,PPV,NPV,ACC and AUC of the combined group in the diagnosis of category4 lesions were 91.67%,95.71%,92.63%,95.12%,94.21% and 0.937 respectively.The constructed logistic regression equation was: Logistic(P)=-9.029+0.087×Age(years)+0.990×Maximum diameter of nodule(cm)+1.804×Nodular edge irregularity+2.313×elastic score≥4 points+3.211×The strain ratio≥2.30.The SEN,SPE,PPV,NPV,ACC and AUC of the model in the diagnosis of BI-RADS category 4 lesions were 90.63%,97.55%,94.98%,95.60%,94.64% and 0.941 respectively.(3)The SEN of the elastic score method in the diagnosis of BI-RADS category 4 lesions was lower than that of other groups,and the difference was statistically significant(P<0.05).The SPE of the conventional ultrasound group in the diagnosis of BI-RADS category 4 lesions was inferior than that of other groups.The SPE of the strain ratio method in the diagnosis of BI-RADS category 4 lesions was lower than that of the elastic score method,and the difference was statistically significant(P<0.05).The ACC of the combined group and the Logistic regression model in the diagnosis of BI-RADS category 4 lesions were higher than those of the conventional ultrasound and the elastic score group,and the difference was statistically significant(P<0.05).The AUC of BI-RADS category 4 lesions was compared by five methods: Logistic regression model>combined group>strain ratio method>conventional ultrasound>elastic score method,the combined group was higher than conventional ultrasound group and elastic score method,the strain ratio method was higher than elastic score method,and the Logistic regression model was higher than strain ratio method,the difference was statistically significant(P<0.05).Conclusion(1)Strain elastography has high diagnostic value for the BI-RADS category4 lesions,and the strain ratio method is superior to the elastic score method.(2)The efficacy of conventional ultrasound combined with elastic score in the diagnosis of BI-RADS category4 lesions is better than that of both alone.(3)The application of stain elastography can reasonably degrade and upgrade the BI-RADS category 4 lesions.