Application Of ARFI In BI-RADS Classification Diagnosis Of Breast Lesions In Dense Breast Backgrounds

Objective: 1.Using acoustic radiation force impulse(ARFI),including acoustic touch tissue imaging(VTI)and acoustic touch tissue quantification(VTQ),to analyze characteristics of benign and malignant lesions,explore the value of ARFI in the qualitative and quantitative diagnosis of breast lesions in dense breast backgrounds,and find out the ARFI indicators which is suitable for diagnosing breast lesions in dense breast backgrounds.2.Exploring the value of BI-RADS classification combined with ARFI for diagnosing breast lesions in dense breast backgrounds.3.Exploring the value of BI-RADS classification for diagnosing breast lesions in dense breast backgrounds.Methods: Retrospective analysis of 123 female patients with mammary gland background underwent surgery in the Cancer Hospital of China Medical University(Liaoning Provincial Cancer Hospital)from October 2017 to October 2019.Patients age were 49.9±9.8 years(Range 25-80);Lesions average diameter were 1.9 ± 0.9 cm(range0.6-4.5).Mammography was performed using a Dimension selenia mammography camera from Hologic,USA.Patients were seated or upright and the cranio-caudal view and the medio-lateral oblique view were taken.Conventional ultrasound and ARFI inspections used Siemens S2000 ultrasound diagnostic equipment,equipped with elastic imaging software and 9L4 linear array high-frequency probe which the frequency was4-9MHz.ARFI included VTI and VTQ imaging methods.These images were independently produced by an ultrasound physician with more than 10 years of experience.These patients underwent mammography for breasts gland density typing.They also received conventional ultrasound examinations for BI-RADS classification.VTI examinations did VTI scoring and calculated the area ratio of VTI image to two-dimensional ultrasound image lesion area.VTQ examinations measured shear wave velocity of the inner and the boundary of the lesions.Finally we calculated the ratio of shear wave velocity of the inner and the boundary of the lesions.BI-RADS classification was according to the fifth edition of the BI-RADS classification standard;a score of less than 4b were judged as benign lesions and others were judged as malignant lesions.We used Tozaki four-point VTI scoring method;a score of less than 4a were judged as benign lesions and others were judged as malignant lesions.IBM SPSS 20.0 statistical software and Med Calc statistical software were used to statistical analysis.Results : 1.The sensitivity,specificity,accuracy,positive predictive value and negative predictive value of conventional ultrasound for diagnosing malignant breast lesions in dense breast backgrounds according to Breast Imaging-Reporting and Data System(BI-RADS)were 89.0%,76.2%,84.7%,88.0%and 78.0% respectively.2.The sensitivity,specificity,accuracy,positive predictive value,and negative predictive value of the VTI score for diagnosing malignant breast lesions in dense breast backgrounds were 78.1%,90.5%,82.3%,94.1%,and 67.9%,respectively.In dense breast backgrounds,the optimal cutoff value of AR for diagnosing malignant breast lesions was 1.2 and the sensitivity,specificity,accuracy,positive predictive value and negative predictive value was 76.8%,88.1 %,79.0%,90.0% and 64.8% respectively.3.In dense breast backgrounds,the optimal cutoff value of shear wave velocity of the inner(SWVi)for diagnosing malignant breast lesions was 5.7m /s and the sensitivity,specificity,accuracy,positive predictive value and negative predictive value was 79.2%,90.5%,83.1%,92.9% and 70.4%respectively.In dense breast backgrounds,the optimal cutoff value of shear wave velocity of the boundary(SWVb)for diagnosing malignant breast lesions was 2.6m/s and the sensitivity,specificity,accuracy,positive predictive value and negative predictive value was 71.6%,69.0%,71.8%,92.9% and 70.4% respectively.In dense breast backgrounds,the optimal cutoff value of lesions to parenchyma ratio(LPR)for diagnosing malignant breast lesions was 1.6 and the sensitivity,specificity,accuracy,positive predictive value and negative predictive value was 75.6%,76.2%,75.8%,86.1%,and 61.5% respectively.4.In dense breast backgrounds,the AR value of the breast malignant lesions group was1.46 ± 0.42,the AR value of the benign breast lesions group was 0.99 ± 0.19;the SWVi of the breast malignant lesions group was 7.55 ± 2.18m/s and The SWVi of the benign lesions group was 3.40 ± 1.83m/s;the SWVb value of the breast malignant lesions group was 3.35 ± 1.36m/s and the SWVb of the benign lesions group was 2.39 ± 0.87m/s;The LPR value of the breast malignant lesions group was 2.50 ± 1.09 and the LPR value of the benign lesions group was 1.55 ± 1.01;The differences between the four ARFI indexes for diagnosing benign and malignant breast lesions in the dense breasts were statistically significant(P <0.05).5.The sensitivity,specificity,accuracy,positive predictive value and negative predictive value of BI-RADS combined with VTI for diagnosing malignant breast lesions in dense breast backgrounds were 87.7%?92.9%?89.5%?96.0% and 79.6%,respectively.6.The sensitivity,specificity,accuracy,positive predictive value and negative predictive value of BI-RADS combined with VTQ for diagnosing malignant breast lesions in dense breast backgrounds were 93.3% ? 83.3% ? 90.3% ? 91.6% and 87.5%,respectively.7.The sensitivity,specificity,accuracy,positive predictive value and negative predictive value of BI-RADS combined with VTI and VTQ for diagnosing malignant breast lesions in dense breast backgrounds were 97.6%,90.5%,95.2%,95.2%,and 95.0%,respectively.Conclusion:All ARFI indicators can be used for diagnosing breast lesions in dense breast backgrounds,of which the VTI score and SWVi are the best diagnostic values;Compared with conventional ultrasound BI-RADS classification,BI-RADS classification combined VTI and VTQ is a better method to better identify difficult lesions in dense breast backgrounds.