Abstract:Background: The high false-positive rates of US BI-RADS category 3-4a breast lesions leads to excessive biopsies of many benign lesions, our aim was to investigate the diagnostic performance achieved by adding a maximum elasticity (Emax) of shear-wave elastography (SWE) to ultrasound (US) to evaluate US Breast Imaging Reporting and Data System (BI-RADS) category 3-4a breast lesions using conservative and aggressive approaches. We explored the capacity of using this method to avoid unnecessary biopsies without increasing the probability of missing breast cancers. Methods: A total of 123 breast lesions of 120 patients classified as BI-RADS category 3 or 4a were enrolled from January 2019 to December 2019. The US features were evaluated according to the US BI-RADS lexicon. The maximum diameter measured on the US was defined as the size of the lesion. The Emax was assessed by SWE, and the average Emax of breast lesions on two images were calculated and recorded as the final maximum Young's modulus. The diagnostic performance of the combined B-mode US and SWE approach for BI-RADS category 3-4a breast lesions was tested using a conservative approach and an aggressive approach. In the conservative approach, the lesions were downgraded with Emax of 30 kPa or less and upgraded with Emax of 160 kPa or more. In the aggressive approach, the lesions were downgraded with Emax of 80 kPa or less and upgraded with Emax of 160 kPa or more. Pathologic results were defined as the reference standard. Results: Among all 123 breast lesions, there were 60 lesions classified as BI-RADS category 3 and 63 lesions classified as BI-RADS category 4a. Compared to the B-mode US, the sensitivity, specificity; positive predictive value (PPV), negative predictive value (NPV), accuracy, and area under the receiver operating characteristic (ROC) curve (AUC) of the combined B-mode US and SWE with a conservative approach changed from 88.9% to 94.4%, 55.2% to 60.0%, 25.4% to 28.8%, 96.7% to 98.4%, 60.2% to 65.0%, and 0.721 to 0.772, respectively. The specificity, PPV, and accuracy of combined B-mode US and SWE with an aggressive approach increased from 55.2% to 72.4%, 25.4% to 29.3%, and 60.2% to 71.5%, respectively, but this was accompanied with decreases in the sensitivity from 88.9% to 66.7%, the NPV from 96.7% to 92.7%, and the AUC from 0.721 to 0.695. Conclusions: The addition of SWE improves the diagnostic performance of breast US. Adding the diagnostic criteria of SWE to the BI-RADS assessment of B-mode US, downgrading the lesions with Emax 30 kPa or less, and upgrading the lesions with Emax 160 kPa or more helped discriminate low suspicion lesions from benign lesions in order to decrease false-positive findings and avoid missing cancer diagnosis.

The dynamin A ring complex: molecular organization and nucleotide‐dependent conformational changes

Artificial intelligence-based, semi-automated segmentation for the extraction of ultrasound-derived radiomics features in breast cancer: a prospective multicenter study

Clinical value of radiomics and machine learning in breast ultrasound: a multicenter study for differential diagnosis of benign and malignant lesions

Management of breast lesions seen on US images: dual-model radiomics including shear-wave elastography may match performance of expert radiologists

US-based radiomics analysis of different machine learning models for differentiating benign and malignant BI-RADS 4A breast lesions

Prediction of the Malignancy of a Breast Lesion Detected on Breast Ultrasound: Radiomics Applied to Clinical Practice

A New Application of Multimodality Radiomics Improves Diagnostic Accuracy of Nonpalpable Breast Lesions in Patients with Microcalcifications-Only in Mammography

Machine learning can reliably predict malignancy of breast lesions based on clinical and ultrasonographic features

Machine Learning Based on Automated Breast Volume Scanner (ABVS) Radiomics for Differential Diagnosis of Benign and Malignant BI‐RADS 4 Lesions

Radiomics and Machine Learning with Multiparametric Breast MRI for Improved Diagnostic Accuracy in Breast Cancer Diagnosis

Multimodal Ultrasound Imaging in Breast Imaging-Reporting and Data System 4 Breast Lesions: A Prediction Model for Malignancy

Applications of machine-learning algorithms for prediction of benign and malignant breast lesions using ultrasound radiomics signatures: A multi-center study

A Machine Learning-Based Unenhanced Radiomics Approach to Distinguishing Between Benign and Malignant Breast Lesions Using T2-Weighted and Diffusion-Weighted MRI.

Differential Diagnosis of B-mode Ultrasound Breast Imaging Reporting and Data System Category 3-4A Lesions in Conjunction with Shear-Wave Elastography Using Conservative and Aggressive Approaches

Ultrasound-based radiomics analysis for differentiating benign and malignant breast lesions: From static images to CEUS video analysis

Automated multiparametric localization of prostate cancer based on B-mode, shear-wave elastography, and contrast-enhanced ultrasound radiomics

Quantitative Evaluation of Peripheral Tissue Elasticity for Ultrasound-Detected Breast Lesions.

Machine Learning Applied to Pre-Operative Computed-Tomography-Based Radiomic Features Can Accurately Differentiate Uterine Leiomyoma from Leiomyosarcoma: A Pilot Study

Diagnostic value of shear wave elastography combined with super microvascular imaging for BI-RADS 3-5 nodules

Radiomics Analysis of Breast Lesions in Combination with Coronal Plane of ABVS and Strain Elastography

Machine learning for multi-parametric breast MRI: radiomics-based approaches for lesion classification