Application of sound touch elastography combined with contrast-enhanced ultrasound in optimizing the diagnosis of BI-RADS 4 breast lesions

Objective To evaluate the application of sound touch elastography(STE) combined with contrast-enhanced ultrasound(CEUS) in optimizing the diagnosis of BI-RADS 4 breast lesions.Methods A total of 91 patients with 93 breast lesions in four BI-RADS types, who were admitted to the Affiliated Hospital of Xuzhou Medical University from December 2020 to June 2022 were enrolled and their clinical data were retrospectively analyzed.The patients underwent conventional ultrasound, STE and CEUS examinations.The elastic modulus of the lesion(A) and its surrounding area(shell 2.0 mm, S)were measured(A_mean, A_max,S_mean,and S_max).According to STE and the prediction CEUS models, the classification of BI-RADS was optimized.Then, ROC curves were plotted, using pathological results as the gold standard. The diagnostic efficiencies of STE, CEUS and both in optimizing BI-RADS classification were compared.Results Among the 93 lesions, 64 were benign and 29 were malignant. The values of A_mean,A_max,S_mean and S_maxin malignant lesions were higher than those in benign breast lesions, with statistical differences(all P<0.05). ROC curves showed that S_max had the highest diagnostic efficiency, at the cut-off value of 106.51 kPa. The sensitivity, specificity, accuracy and AUC of S_max in optimizing BI-RADS classification was75.86%, 87.50%, 83.87% and 0.807, respectively. The sensitivity, specificity, accuracy and AUC of CEUS in optimizing BI-RADS classification was 82.76%,84.38%,83.87% and 0.836, respectively. The sensitivity, specificity, accuracy and AUC of both STE and CEUS in optimizing BI-RADS classification was 89.66%,93.75%,92.47% and 0.917, respectively, which were higher than those of single optimized classification(P<0.05).Conclusions The optimization of BI-RADS classification by STE and CEUS can significantly improve the diagnostic efficiency of ultrasound in distinguishing benign and malignant breast lesions and avoid unnecessary puncture biopsies.