Search

Convolutional neural networks (CNNs) have been successfully applied in the computer-aided ultrasound diagnosis for breast cancer. Up to now, several CNN-based methods have been proposed. However, most of them consider tumor localization and classification as two separate steps, rather than performing them simultaneously. Besides, they suffer from the limited diagnosis information in the B-mode ultrasound (BUS) images. In this study, we develop a novel network ResNet-GAP that incorporates both localization and classification into a unified procedure. To enhance the performance of ResNet-GAP, we leverage stiffness information in the elastography ultrasound (EUS) modality by collaborative learning in the training stage. Specifically, a dual-channel ResNet-GAP network is developed, one channel for BUS and the other for EUS. In each channel, multiple class activity maps (CAMs) are generated using a series of convolutional kernels of different sizes. The multi-scale consistency of the CAMs in both channels are further considered in network optimization. Experiments on 264 patients in this study show that the newly developed ResNet-GAP achieves an accuracy of 88.6%, a sensitivity of 95.3%, a specificity of 84.6%, and an AUC of 93.6% on the classification task, and a 1.0NLF of 87.9% on the localization task, which is better than some state-of-the-art approaches.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and ultrasound (US), which are two common modalities for clinical breast tumor diagnosis besides Mammograms, can provide different and complementary information for the same tumor regions. Although many machine learning methods have been proposed for breast tumor classification based on either single modality, it remains unclear how to further boost the classification performance by utilizing paired multi-modality information with different dimensions. In this paper, we propose MRI-US multi-modality network (MUM-Net) to classify breast tumor into different subtypes based on 3D MR and 2D US images. The key insight of MUM-Net is that we explicitly distill modality-agnostic features for tumor classification. Specifically, we first adopt a discrimination-adaption module to decompose features into modality-agnostic and modality-specific ones with min-max training strategies. Then, we propose a feature fusion module to increase the compactness of the modality-agnostic features by utilizing an affinity matrix with nearest neighbour selection. We build a paired MRI-US breast tumor classification dataset containing 502 cases with three clinical indicators to validate the proposed method. In three tasks including lymph node metastasis, histological grade and Ki-67 level, MUM-Net achieves AUC scores of 0.8581, 0.8965 and 0.8577, outperforming other counterparts which are based on single task or single modality by a wide margin. In addition, we find that the extracted modality-agnostic features can help the network focus on the tumor regions in both modalities.

Objective This study aimed to incorporate clinicopathological, sonographic, and mammographic characteristics to construct and validate a nomogram model for predicting disease-free survival (DFS) in patients with triple-negative breast cancer (TNBC). Methods Patients diagnosed with TNBC at our institution between 2011 and 2015 were retrospectively evaluated. A nomogram model was generated based on clinicopathological, sonographic, and mammographic variables that were associated with 1-, 3-, and 5-year DFS determined by multivariate logistic regression analysis in the training set. The nomogram model was validated according to the concordance index (C-index) and calibration curves in the validation set. Results A total of 636 TNBC patients were enrolled and divided into training cohort (n = 446) and validation cohort (n = 190). Clinical factors including tumor size > 2 cm, axillary dissection, presence of LVI, and sonographic features such as angular/spiculated margins, posterior acoustic shadows, and presence of suspicious lymph nodes on preoperative US showed a tendency towards worse DFS. The multivariate analysis showed that no adjuvant chemotherapy (HR = 6.7, 95% CI: 2.6, 17.5, p < 0.0005), higher axillary tumor burden (HR = 2.7, 95% CI: 1.0, 7.1, p = 0.045), and ≥ 3 malignant features on ultrasound (HR = 2.4, CI: 1.1, 5.0, p = 0.021) were identified as independent prognostic factors associated with poorer DFS outcomes. In the nomogram, the C-index was 0.693 for the training cohort and 0.694 for the validation cohort. The calibration plots also exhibited excellent consistency between the nomogram-predicted and actual survival probabilities in both the training and validation cohorts. Conclusions Clinical variables and sonographic features were correlated with the prognosis of TNBCs. The nomogram model based on three variables including no adjuvant chemotherapy, higher axillary tumor load, and more malignant sonographic features showed good predictive performance for poor survival outcomes of TNBC. Key Points • The absence of adjuvant chemotherapy, heavy axillary tumor load, and malignant-like sonographic features can predict DFS in patients with TNBC. • Mammographic features of TNBC could not predict the survival outcomes of patients with TNBC. • The nomogram integrating clinicopathological and sonographic characteristics is a reliable predictive model for the prognostic outcome of TNBC.

The sonographic appearance of benign and malignant breast nodules overlaps to some extent, and we aimed to assess the performance of the Gail model as an adjunctive tool to ultrasound (US) Breast Imaging Reporting and Data System (BI-RADS) for predicting the malignancy of nodules.From 2018 to 2019, 2607 patients were prospectively enrolled by 35 health care facilities. An individual breast cancer risk was assessed by the Gail model. Based on B-mode US, color Doppler, and elastography, all nodules were evaluated according to the fifth edition of BI-RADS, and these nodules were all confirmed later by pathology.We demonstrated that the Gail model, age, tumor size, tumor shape, growth orientation, margin, contour, acoustic shadowing, microcalcification, presence of duct ectasia, presence of architectural distortion, color Doppler flow, BI-RADS, and elastography score were significantly related to breast cancer (all p0.001). The sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and area under the curve (AUC) for combining the Gail model with the BI-RADS category were 95.6%, 91.3%, 85.0%, 97.6%, 92.8%, and 0.98, respectively. Combining the Gail model with the BI-RADS showed better diagnostic efficiency than the BI-RADS and Gail model alone (AUC 0.98 vs 0.80, p0.001; AUC 0.98 vs 0.55, p0.001) and demonstrated a higher specificity than the BI-RADS (91.3% vs 59.4%, p0.001).The Gail model could be used to differentiate malignant and benign breast lesions. Combined with the BI-RADS category, the Gail model was adjunctive to US for predicting breast lesions for malignancy. For the diagnosis of malignancy, more attention should be paid to high-risk patients with breast lesions.

Background Human epidermal growth factor receptor2+ subtype breast cancer has a high degree of malignancy and a poor prognosis. The aim of this study is to develop a prediction model for the human epidermal growth factor receptor2+ subtype (non-luminal) of breast cancer based on the clinical and ultrasound features related with estrogen receptor, progesterone receptor, and human epidermal growth factor receptor2. Methods We collected clinical data and reviewed preoperative ultrasound images of enrolled breast cancers from September 2017 to August 2020. We divided the data into in three groups as follows. Group I: estrogen receptor ± , Group II: progesterone receptor ± and Group III: human epidermal growth factor receptor2 ± . Univariate and multivariate logistic regression analyses were used to analyze the clinical and ultrasound features related with biomarkers among these groups. A model to predict human epidermal growth factor receptor2+ subtype was then developed based on the results of multivariate regression analyses, and the efficacy was evaluated using the area under receiver operating characteristic curve, accuracy, sensitivity, specificity. Results The human epidermal growth factor receptor2+ subtype accounted for 138 cases (11.8%) in the training set and 51 cases (10.1%) in the test set. In the multivariate regression analysis, age ≤ 50 years was an independent predictor of progesterone receptor + (p = 0.007), and posterior enhancement was a negative predictor of progesterone receptor + (p = 0.013) in Group II; palpable axillary lymph node, round, irregular shape and calcifications were independent predictors of the positivity for human epidermal growth factor receptor-2 in Group III (p = 0.001, p = 0.007, p = 0.010, p Conclusions Our study established a model to predict the human epidermal growth factor receptor2-positive subtype with moderate performance. And the results demonstrated that clinical and ultrasound features were significantly associated with biomarkers.

Sonographic features are associated with pathological and immunohistochemical characteristics of triple-negative breast cancer (TNBC). To predict the biological property of TNBC, the performance using quantitative high-throughput sonographic feature analysis was compared with that using qualitative feature assessment. We retrospectively reviewed ultrasound images, clinical, pathological, and immunohistochemical (IHC) data of 252 female TNBC patients. All patients were subgrouped according to the histological grade, Ki67 expression level, and human epidermal growth factor receptor 2 (HER2) score. Qualitative sonographic feature assessment included shape, margin, posterior acoustic pattern, and calcification referring to the Breast Imaging Reporting and Data System (BI-RADS). Quantitative sonographic features were acquired based on the computer-aided radiomics analysis. Breast cancer masses were manually segmented from the surrounding breast tissues. For each ultrasound image, 1688 radiomics features of 7 feature classes were extracted. The principal component analysis (PCA), least absolute shrinkage and selection operator (LASSO), and support vector machine (SVM) were used to determine the high-throughput radiomics features that were highly correlated to biological properties. The performance using both quantitative and qualitative sonographic features to predict biological properties of TNBC was represented by the area under the receiver operating characteristic curve (AUC). In the qualitative assessment, regular tumor shape, no angular or spiculated margin, posterior acoustic enhancement, and no calcification were used as the independent sonographic features for TNBC. Using the combination of these four features to predict the histological grade, Ki67, HER2, axillary lymph node metastasis (ALNM), and lymphovascular invasion (LVI), the AUC was 0.673, 0.680, 0.651, 0.587, and 0.566, respectively. The number of high-throughput features that closely correlated with biological properties was 34 for histological grade (AUC 0.942), 27 for Ki67 (AUC 0.732), 25 for HER2 (AUC 0.730), 34 for ALNM (AUC 0.804), and 34 for LVI (AUC 0.795). High-throughput quantitative sonographic features are superior to traditional qualitative ultrasound features in predicting the biological behavior of TNBC. • Sonographic appearances of TNBCs showed a great variety in accordance with its biological and clinical characteristics. • Both qualitative and quantitative sonographic features of TNBCs are associated with tumor biological characteristics. • The quantitative high-throughput feature analysis is superior to two-dimensional sonographic feature assessment in predicting tumor biological property.

Background An accurate preoperative assessment of cervical lymph node metastasis (LNM) is important for choosing an optimal therapeutic strategy for papillary thyroid carcinoma (PTC) patients. This study aimed to develop and validate two ultrasound (US) nomograms for the individual prediction of central and lateral compartment LNM in patients with PTC. Methods A total of 720 PTC patients from 3 institutions were enrolled in this study. They were categorized into a primary cohort, an internal validation, and two external validation cohorts. Radiomics features were extracted from conventional US images. LASSO regression was used to select optimized features to construct the radiomics signature. Two nomograms integrating independent clinical variables and radiomics signature were established with multivariate logistic regression. The performance of the nomograms was assessed with regard to discrimination, calibration, and clinical usefulness. Results The radiomics scores were significantly higher in patients with central/lateral LNM. A radiomics nomogram indicated good discrimination for central compartment LNM, with an area under the curve (AUC) of 0.875 in the training set, the corresponding value in the validation sets were 0.856, 0.870 and 0.870, respectively. Another nomogram for predicting lateral LNM also demonstrated good performance with an AUC of 0.938 and 0.905 in the training and internal validation cohorts, respectively. The AUC for the two external validation cohorts were 0.881 and 0.903, respectively. The clinical utility of the nomograms was confirmed by the decision curve analysis. Conclusion The nomograms proposed here have favorable performance for preoperatively predicting cervical LNM, hold promise for optimizing the personalized treatment, and might greatly facilitate the decision-making in clinical practice.

The aim of the work described here was to develop an ultrasound (US) image-based deep learning model to reduce the rate of malignancy among breast lesions diagnosed as category 4A of the Breast Imaging-Reporting and Data System (BI-RADS) during the pre-operative US examination. A total of 479 breast lesions diagnosed as BI-RADS 4A in pre-operative US examination were enrolled. There were 362 benign lesions and 117 malignant lesions confirmed by postoperative pathology with a malignancy rate of 24.4%. US images were collected from the database server. They were then randomly divided into training and testing cohorts at a ratio of 4:1. To correctly classify malignant and benign tumors diagnosed as BI-RADS 4A in US, four deep learning models, including MobileNet, DenseNet121, Xception and Inception V3, were developed. The performance of deep learning models was compared using the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Meanwhile, the robustness of the models was evaluated by five-fold cross-validation. Among the four models, the MobileNet model turned to be the optimal model with the best performance in classifying benign and malignant lesions among BI-RADS 4A breast lesions. The AUROC, accuracy, sensitivity, specificity, PPV and NPV of the optimal model in the testing cohort were 0.897, 0.913, 0.926, 0.899, 0.958 and 0.784, respectively. About 14.4% of patients were expected to be upgraded to BI-RADS 4B in US with the assistance of the MobileNet model. The deep learning model MobileNet can help to reduce the rate of malignancy among BI-RADS 4A breast lesions in pre-operative US examinations, which is valuable to clinicians in tailoring treatment for suspicious breast lesions identified on US.

Non-invasive assessment of the risk of lymph node metastasis (LNM) in patients with papillary thyroid carcinoma (PTC) is of great value for the treatment option selection. The purpose of this paper is to develop a transfer learning radiomics (TLR) model for preoperative prediction of LNM in PTC patients in a multicenter, cross-machine, multi-operator scenario. Here we report the TLR model produces a stable LNM prediction. In the experiments of cross-validation and independent testing of the main cohort according to diagnostic time, machine, and operator, the TLR achieves an average area under the curve (AUC) of 0.90. In the other two independent cohorts, TLR also achieves 0.93 AUC, and this performance is statistically better than the other three methods according to Delong test. Decision curve analysis also proves that the TLR model brings more benefit to PTC patients than other methods., A non-destructive and efficient method for predicting the risk of lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC) patients is highly needed. Here, the authors develop a transfer learning radiomics model for preoperative prediction of LNM in patients with PTC in a multicenter scenario.

Purpose To determine the overall performance of contrast-enhanced ultrasound (CEUS) in differentiating between benign and malignant breast lesions and in predicting the pathologic response to neoadjuvant chemotherapy (NAC) in patients with breast cancer (BC). Materials and Methods Articles published up to April 2019 were systematically searched in Medline, Web of Science, and China National Knowledge Infrastructure. The sensitivities and specificities across studies, the calculations of positive and negative likelihood ratios (LR+ and LR−), diagnostic odds ratio (OR), and constructed summary receiver operating characteristic curves were determined. Methodologic quality was assessed using the QUADAS (Quality Assessment of Diagnostic Accuracy Studies) tool. Subgroup analyses and metaregression were performed on prespecified study-level characteristics. Results Fifty-one studies involving 4875 patients with 5246 breast lesions and 10 studies involving 462 patients with BC receiving NAC were included. Methodologic quality was relatively high, and no publication bias was detected. The overall sensitivity, specificity, diagnostic OR, LR+, and LR− for CEUS were 0.88 (95% confidence interval [CI], 0.86-0.89), 0.82 (95% CI, 0.80-0.83), 30.55 (95% CI, 21.40-43.62), 4.29 (95% CI, 3.51-5.25), and 0.16 (95% CI, 0.13-0.21), respectively, showing statistical heterogeneity. Multivariable metaregression analysis showed contrast mode to be the most significant source of heterogeneity. The overall sensitivity, specificity, LR+, LR, and diagnostic OR of CEUS imaging in predicting the overall pathologic response to NAC in patients with BC were 0.89 (95% CI, 0.83-0.93), 0.83 (95% CI, 0.78-0.88), 4.49 (95% CI, 3.04-6.62), 0.16 (95% CI, 0.10-0.24,), and 32.21 (95% CI, 16.74-62.01), respectively, showing mild heterogeneity. Conclusion Our data confirmed the excellent performance of breast CEUS in differentiating between benign and malignant breast lesions as well as pathologic response prediction in patients with BC receiving NAC.

OBJECTIVES To investigate the correlation between ultrasound (US) appearances of invasive breast cancers and tumor proliferation and invasiveness measured according to the histologic grade, Ki-67 expression, axillary lymph node metastasis (ALNM), and lymphovascular invasion (LVI). METHODS This study evaluated 676 patients who underwent primary surgical treatment of invasive breast cancers. The preoperative US reports and postoperative pathologic and immunohistochemical results of the patients were retrospectively reviewed. Ultrasound characteristics were evaluated according to the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) lexicon. Logistic regression analyses were used to identify independent predictive US features that were correlated with tumor proliferation and invasiveness of breast cancers. Odds ratios (ORs) were calculated. RESULTS Posterior acoustic enhancement and calcifications on US images were independent predictive factors of a higher histologic grade and a higher Ki-67 level (OR, 1.69-6.54; P

Background and Objectives: The breast undergoes extensive physiologic changes during pregnancy/lactation. We aimed to review the ultrasound (US) imaging characteristics of breast lesions during pregnancy/lactation and to demonstrate ultrasonography as an excellent imaging modality in this patient population. Materials and Methods: We performed a retrospective study involving 195 patients with 206 pathologically confirmed breast lesions during pregnancy/lactation over the period of January 2010 to December 2018; 51 were diagnosed with breast cancer, including 50 invasive ductal carcinomas and 1 low malignant potential phyllodes tumor, whereas 144 were diagnosed with 155 benign lesions, including 103 mastitis/abscesses, 45 fibroadenomas, 2 intraductal papillomas, 2 sclerosing lesions, and 3 benign phyllodes tumors. The patients' age, lesion diameter, and US characteristics were analyzed. Results: The breast lesions of patients during pregnancy/lactation were compared with those during nonpregnancy/lactation. Patients with breast cancer or fibroadenomas during pregnancy/lactation were younger than those during nonpregnancy/lactation. The average lesion diameter was significantly higher among pregnant/lactating patients compared with controls of childbearing age for fibroadenomas, but not for malignant lesions. The fibroadenomas and mastitis/abscesses during pregnancy/lactation usually have higher BI-RADS categories than those during nonpregnancy/lactation. Conclusions: The imaging features of breast cancer during pregnancy/lactation did not differ much from those of nonpregnancy/lactation; however, some benign lesions had suspicious sonographic features, and US-guided core biopsies were necessary for differentiating benign from malignant lesions.

Objective To assess the performance of conventional ultrasound (US) combined with strain elastography (SE) in the Breast Imaging Reporting and Data System (BI-RADS) category 4 lesions on mammography. Methods Women with breast lesions identified as having mammography BI-RADS 4 lesions and underwent US examination were included in China. US features and US BI-RADS assessment were recorded in real-time and prospectively reported. The pathological result was referred to as the gold standard. The performance of US in the mammographic BI-RADS category 4 lesions was evaluated. Diagnostic performances of US BI-RADS, SE and combined both were compared. Results 751 women with 751 breast lesions classified as mammographic BI-RADS category 4 were included. For mammographic findings, 530 (70.6%) were true-positive and 221 (29.4%) were false-positive. Conventional US achieved higher positive predictive value (PPV) than mammography (78.5% vs 70.6%, p=0.001). The specificity increased from 34.4% to 47.1% (p Conclusions Conventional US combined with SE is a helpful tool for the non-invasive examination of breast lesions classified as BI-RADS category 4 on mammography. It helped increase the PPV and had the potential to avoid unnecessary biopsies of BI-RADS category 4 lesions detected on mammography. Micro Abstract We assessed the performance of conventional ultrasound (US) combined with strain elastography (SE) in the Breast Imaging Reporting and Data System (BI-RADS) category 4 lesions on mammography. 751 women with 751 breast lesions classified as mammographic BI-RADS category 4 were included. Conventional US combined with SE helped increase the PPV and had the potential to avoid unnecessary biopsies of BI-RADS category 4 lesions detected on mammography.

Background: To compare the abilities of ultrasonography (US) and Computed Tomography (CT) to identify calcifications and to predict probability of malignancy for Papillary Thyroid Carcinoma (PTC) and Papillary Thyroid Microcarcinoma (PTMC). Methods: We reviewed 1008 cases of PTC/PTMC with calcifications reported by pre-operative US, CT, or post-operative pathology. The size of the thyroid nodule was obtained from the US report and the maximum diameter (d) was documented. According to the nodule size (d), the PTC and PTMC groups were each divided into two subgroups, as follows: large PTC group (d ≥ 2 cm), small PTC group (1 cm < d < 2 cm), large PTMC group (0.6 cm ≤ d ≤ 1 cm), and small PTMC group (d < 0.6 cm). Results: In the 1008 patients, the ratio of females to males was 2.29 and the mean age was 40.9 years (standard deviation: 11.7 years). Of the 1008 records, 92.8% were found to have calcifications according to the US report, while 50.4% showed calcifications according to the CT report. This difference between US and CT reports was statistically significant (p < 0.0005). The percentages of US reports showing calcifications were similar for all four PTC and PTMC subgroups (93.7%, 94.3%, 92.1%, and 85.1%, respectively; p = 0.052), while the percentages of CT reports showing calcifications were significantly different among the PTC and PTMC subgroups (62.3%, 52.2%, 45.4%, and 31.3%, respectively; p < 0.0005). As for the prediction of malignancy, US was superior to CT in all four subgroups (large PTC group: 97.1% vs. 54.1%, small PTC group: 94.8% vs. 42.9%, large PTMC group: 97.2% vs. 32.0%, small PTMC group: 95.5% vs. 14.9%; p < 0.0005 for all pairwise comparisons). No significant difference was observed in terms of the ability of US to predict the malignancy of PTC versus PTMC (p = 0.31), while CT showed significant superiority in diagnosing PTC versus PTMC (p < 0.0005). The predictive value of CT for PTC declined as the nodule size decreased (p < 0.05 for all pairwise comparisons). Conclusion: Our results showed that US detected calcifications and predicted the malignancy of all nodule sizes of thyroid papillary carcinoma equally well, while the performance of CT declined with the reduction of nodule size.

Purpose Due to the low contrast, blurry boundaries, and large amount of shadows in breast ultrasound (BUS) images, automatic tumor segmentation remains a challenging task. Deep learning provides a solution to this problem, since it can effectively extract representative features from lesions and the background in BUS images. Methods A novel automatic tumor segmentation method is proposed by combining a dilated fully convolutional network (DFCN) with a phase-based active contour (PBAC) model. The DFCN is an improved fully convolutional neural network with dilated convolution in deeper layers, fewer parameters, and batch normalization techniques; and has a large receptive field that can separate tumors from background. The predictions made by the DFCN are relatively rough due to blurry boundaries and variations in tumor sizes; thus, the PBAC model, which adds both region-based and phase-based energy functions, is applied to further improve segmentation results. The DFCN model is trained and tested in dataset 1 which contains 570 BUS images from 89 patients. In dataset 2, a 10-fold support vector machine (SVM) classifier is employed to verify the diagnostic ability using 460 features extracted from the segmentation results of the proposed method. Results Advantages of the present method were compared with three state-of-the-art networks; the FCN-8s, U-net, and dilated residual network (DRN). Experimental results from 170 BUS images show that the proposed method had a Dice Similarity coefficient of 88.97 ± 10.01%, a Hausdorff distance (HD) of 35.54 ± 29.70 pixels, and a mean absolute deviation (MAD) of 7.67 ± 6.67 pixels, which showed the best segmentation performance. In dataset 2, the area under curve (AUC) of the 10-fold SVM classifier was 0.795 which is similar to the classification using the manual segmentation results. Conclusions The proposed automatic method may be sufficiently accurate, robust, and efficient for medical ultrasound applications.

OBJECTIVES To identify clinicopathologic and ultrasound (US) variables that were associated with a heavy nodal tumor burden, which was defined as 3 or more lymph nodes involved with metastasis to the axilla after invasive breast carcinoma. METHODS With ethical approval, 621 patients with a pathologic diagnosis of invasive breast carcinoma were retrospectively analyzed for clinical, pathologic, and US data. Pathologic findings were ascertained by the final paraffin pathologic analysis. Ultrasound characteristics were evaluated on the basis of the American College of Radiology's Breast Imaging Reporting and Data System (BI-RADS). Univariate and multivariate logistic regression analyses were used to assess the clinicopathologic and US variables that were associated with a heavy nodal tumor burden at the axilla. RESULTS There were 107 cases (17.2%) of invasive breast carcinoma with a heavy tumor burden at the axilla. The independent clinicopathologic variables for a heavy tumor burden at the axilla included a tumor size of 2 to 5 cm (odds ratio [OR], 1.86; P = .036), the presence of lymphovascular invasion (OR, 23.52; P

The accurate localization of nodules in ultrasound images can convey crucial information to support a reliable diagnosis. However, this is usually challenging due to low contrast and image artifacts, especially in thyroid ultrasound images where nodules are relatively small in most cases. To address these problems, in this paper, we propose a joint-training convolutional neural network (CNN) for thyroid nodule localization in ultrasound images. Considering the advantage of the faster region-based CNN (Faster R-CNN) in detecting natural targets, we adopt it as the basic framework. To boost the representative power and noise suppression capability of the network, the attention mechanism module is embedded for adaptive feature refinement along the channel and spatial dimensions. Furthermore, in the training process, we annotate the training set in a novel way, called joint-training annotation, by exploiting the fake foreground (FFG) area around the nodule as a spatial prior constraint to improve the sensitivity to small nodules. Ablation experiments are conducted to verify the effectiveness of our proposed method. The experimental results show that our method outperforms others by a mean average precision (mAP) of 0.93 and achieves an intersection over union (IoU) of 0.9, indicating that the localization results agree well with the ground truth. Furthermore, extended experiments on breast nodule datasets are also conducted to verify the generalizability of the proposed approach. Above all, the proposed algorithm is of considerable significance for accurate thyroid nodule localization in ultrasound images and can be generalized to other types of nodules, thereby providing trustworthy assistance for clinical diagnosis.

Accurate preoperative identification of lateral cervical lymph node metastasis (LNM) is important for decision-making and clinical management of patients with papillary thyroid carcinoma (PTC). The aim of this study was to develop an ultrasound (US)-based radiomic nomogram to preoperatively predict the lateral LNM in PTC patients.In this retrospective study, a total of 886 patients were enrolled and randomly divided into 2 groups. Radiomic features were extracted from the preoperative US images. A radiomic signature was constructed using the least absolute shrinkage and selection operator algorithm in the training set. Multivariate logistic regression was performed to develop the radiomic nomogram, which incorporating the radiomic signature and the selected clinical characteristics. The performance of the nomogram was assessed by its discrimination, calibration, and clinical usefulness in both the training and validation sets.The radiomic signature was significantly associated with the lateral LNM in both cohorts (p0.001). The nomogram that consisted of radiomic signature, US-reported cervical lymph node (CLN) status, and CT-reported CLN status demonstrated good discrimination and calibration in the training and validation sets with an AUC of 0.946 and 0.914, respectively. The decision curve analysis indicated that the radiomic nomogram was worthy of clinical application.The radiomic nomogram proposed here has good performance for noninvasively predicting the lateral LNM and might be used to facilitate clinical decision-making and potentially improve the survival outcome in selected patients.

Breast calcifications indicate the high possibility of malignancy in the radiological assessment of breast lesions. However, it is difficult to detect them from traditional B-mode ultrasound images due to the resolution limit and speckle noise. In this paper, we proposed a novel automatic calcification detection method based on ultrasound radio frequency (RF) signals by quantitative multi-parameter fusion. The proposed method consists of four steps: selecting the region of interest (ROI), extracting multiple features on sliding windows that traverse the entire ROI, classifying the window with or without calcifications using the Adaptive Boosting classifier, and obtaining the detection result by a threshold filter. Experiments were conducted on a database of 130 experienced doctor-proven breast tumors with calcifications. Compared to manual annotation, the proposed method achieved an average accuracy of 88%. The experiments demonstrated that our computerized RF signals feature system was capable of helping radiologists detect tumor calcifications more accurately and provided more guidance for the final decision.

OBJECTIVES To investigate the value of ultrasound (US) feature-based models in predicting the proliferation and invasiveness of invasive breast cancer (IBC) and to compare the performance of models based solely on US features with models that combined US features, patient age, tumor size, and axilla status from US. METHODS With ethical approval, 746 patients with a pathologic diagnosis of IBC were reviewed for preoperative clinical, US, and postoperative pathologic data. The proliferation and invasiveness properties of the IBC included the histologic grade and Ki-67 status and lymphovascular invasion (LVI) and axillary lymph node metastasis (ALNM), respectively. Logistic regression analyses were used to identify independent risk factors for tumor proliferation and invasiveness. RESULTS Posterior echo enhancement, calcification, a tumor size larger than 2 cm, and suspicion of ALNM from axillary US were independent risk factors for a high histologic grade and high Ki-67 expression of IBC (P < .05). A posterior echo shadow, patient age younger than 45 years, and suspicious findings on axillary US imaging were independent variables for predicting the presence of LVI and ALNM in IBC (P < .05). Calcification was the independent factor for predicting LVI (P = .013). The predictive performance of the combined models was improved compared with the US feature-based models, with a higher accuracy rate and negative predictive value. The area under curve of the combined models was also significantly higher than that of the single models (P < .05). CONCLUSIONS Compared with the US feature-based models, the combined models yielded better predictive performance. This may provide a more robust model to predict the tumor biological properties of IBC before surgery.

PURPOSE Digital Breast Imaging Reporting and Data System (BI-RADS) features extracted from ultrasound images are essential in computer-aided diagnosis, prediction, and prognosis of breast cancer. This study focuses on the reproducibility of quantitative high-throughput BI-RADS features in the presence of variations due to different segmentation results, various ultrasound machine models, and multiple ultrasound machine settings. METHODS Dataset 1 consists of 399 patients with invasive breast cancer and is used as the training set to measure the reproducibility of features, while dataset 2 consists of 138 other patients and is a validation set used to evaluate the diagnosis performances of the final reproducible features. Four hundred and sixty high-throughput BI-RADS features are designed and quantized according to BI-RADS lexicon. Concordance Correlation Coefficient (CCC) and Deviation (Dev) are used to assess the effect of the segmentation methods and Between-class Distance (BD) is used to study the influences of the machine models. In addition, the features jointly shared by two methodologies are further investigated on their effects with multiple machine settings. Subsequently, the absolute value of Pearson Correlation Coefficient (Rabs ) is applied for redundancy elimination. Finally, the features that are reproducible and not redundant are preserved as the stable feature set. A 10-fold Support Vector Machine (SVM) classifier is employed to verify the diagnostic ability. RESULTS One hundred and fifty-three features were found to have high reproducibility (CCC > 0.9 & Dev < 0.1) within the manual and automatic segmentation. Three hundred and thirty-nine features were stable (BD < 0.2) at different machine models. Two feature sets shared the same 102 features, in which nine features were highly sensitive to the machine settings. Forty-six features were finally preserved after redundancy elimination. For the validation in dataset 2, the area under curve (AUC) of the 10-fold SVM classifier was 0.915. CONCLUSIONS Three factors, segmentation results, machine models, and machine settings may affect the reproducibility of high-throughput BI-RADS features to various degrees. Our 46 reproducible features were robust to these factors and were capable of distinguishing benign and malignant breast tumors.

OBJECTIVES We aimed to determine whether the integration of shear wave elastography (SWE) with conventional ultrasonography (US) improves diagnostic performance for suspicious thyroid lesions. METHODS For 215 thyroid lesions in 185 patients classified as Thyroid Imaging Reporting and Data System category 4 or 5 according to the findings of conventional US, SWE elasticity indices were automatically calculated. A receiver operating characteristic curve analysis was used to determine the threshold. Thyroid Imaging Reporting and Data System categories were upgraded for high-stiffness nodules and unchanged for low- and normal-stiffness nodules. The diagnostic performances were assessed and compared with histologic findings. Intraobserver and interobserver variability of SWE was assessed. RESULTS Elasticity indices were significantly higher in malignant versus benign nodules (P≤ .001). The minimum elasticity index (cutoff, 40.7 kPa) of the stiffest part combined with conventional US showed the highest area under the curve (0.774; 95% confidence interval, 0.682-0.866) but was not superior to conventional US (0.791; 95% confidence interval, 0.706-0.876; P = .48). Combined with the standard deviation of the elasticity index for the whole lesion (cutoff, 6.8 kPa), US yielded the highest sensitivity (95.5%; P < .001) and lowest specificity (42.1%; P < .001). Sensitivity increased and specificity decreased by adding any other SWE elasticity index. The intraobserver and interobserver reliability of SWE was fair to excellent according to the interclass correlation coefficients, with correlation coefficients of 0.765 to 0.846 (all P < .001). CONCLUSIONS The SWE elasticity indices of malignant thyroid nodules were significantly high. Adding SWE to conventional US did not improve diagnostic performance.

OBJECTIVES The primary objective of this study was to evaluate the difference and agreement between ultrasonography (US) and computed tomography (CT) for identifying calcifications in thyroid nodules. METHODS Data from the medical records of 20,248 patients were reviewed for preoperative diagnostic investigations and postoperative pathologic diagnoses. In total, 5247 records were selected for analysis based on the presence of calcifications reported in any of the following 3 modes: US, CT, and pathologic analysis. All 5247 patients underwent US examinations, whereas 3827 underwent cervical CT examinations. All patients had a postoperative pathologic diagnosis serving as a reference. The value of US for identification of calcifications and prediction of malignancy was analyzed on the basis of the entire cohort of 5247 records, whereas that of CT was based on 3827 records. The agreement between US and CT was analyzed on the basis of the 3827 common records. RESULTS Of the 5247 patients who underwent US, 4855 (92.5%) were found to have calcifications, whereas of the 3827 patients who underwent CT, 2040 (53.3%) were found to have calcifications (P < .0005). Among the 404 cases with calcifications reported by pathologic analysis, the agreement rate between US and pathologic findings was significantly higher than that between CT and pathologic findings (87.9% versus 81.9%, respectively; P = .018). CONCLUSIONS US is more sensitive and accurate than CT for detecting calcifications in thyroid nodules. Hence, US is recommended as the preferred imaging modality for calcification detection in thyroid nodules.

B-mode ultrasound (BUS) imaging is widely used for diagnosis of breast tumors. In recent years, another ultrasound modality, namely ultrasound elastography (UE), has also been successfully applied in clinical practice. Although the combination of bimodal ultrasound imaging can improve the diagnosis accuracy for breast tumors, the single-modal ultrasound-based diagnosis is more popular in clinical practice, especially in the rural areas. Since transfer learning (TL) can transfer knowledge from a source domain to a target domain to help train more effective model, we propose to develop a single-modal ultrasound-based computer-aided diagnosis (CAD) with the transferred information from an additional modality in this work. We propose a projective model (PM) based multilayer kernel extreme learning machine (ML-KELM-PM) algorithm, which performs the parameter transfer approach in classifier to perform TL in CAD. The experimental results on a bimodal ultrasound image dataset show that the proposed ML-KELM-PM algorithm outperforms all the compared algorithms for the single-modal ultrasound-based CAD for breast tumors.

In this study, we aimed to evaluate the clinical and pathological factors that associated with sonographic appearances of triple-negative (TN) invasive breast carcinoma. With the ethical approval, 560 patients who were pathologically confirmed as invasive breast carcinoma were reviewed for ultrasound, clinical, and pathological data. Logistic regression analysis was used to identify the typical sonographic features for TN invasive breast carcinomas. The effect of clinical and pathological factors on the sonographic features of TN invasive breast carcinoma was studied. There were 104 cases of TN invasive breast carcinoma. The independent sonographic features for the TN subgroup included regular shape (odds ratio, OR = 1.73, p = 0.033), no spiculated/angular margin (OR = 2.09, p = 0.01), posterior acoustic enhancement (OR = 2.09, p = 0.004), and no calcifications (OR = 2.11, p = 0.005). Higher pathological grade was significantly associated with regular tumor shape of TN breast cancer (p = 0.012). Higher Ki67 level was significantly associated with regular tumor shape (p = 0.023) and absence of angular/spiculated margin (p = 0.005). Higher human epidermal growth factor receptor 2 (HER2) score was significantly associated with the presence of calcifications (p = 0.033). We conclude that four sonographic features are associated with TN invasive breast carcinoma. Heterogeneity of sonographic features was associated with the pathological grade, Ki67 proliferation level and HER2 score of TN breast cancers.

Background Ultrasound shear-wave elastography (SWE) may increase specificity of breast lesion assessment with ultrasound, but elasticity measurements may change with transducer orientation, defined as anisotropy. In this study, we aimed to observe the anisotropy of SWE of breast lesions, and its correlation with clinical and histopathological findings. Methods This retrospective study was approved by institutional review board. From June 2014 to June 2015, a total of 276 women (mean age, 48.75 ± 12.12 years) with 276 breast lesions (174 malignant, 102 benign) were enrolled for conventional ultrasound and SWE before surgical excision. Elasticity modulus in the longest diameter and orthogonal diameter were recorded, including maximum elasticity (Emax), mean elasticity (Emean), standard deviation (Esd) and ratio between mean elasticity of lesion and normal fatty tissue (Eratio). Anisotropy coefficients including anisotropic difference (AD) and anisotropy factors (AF) were calculated, and correlations with malignancy, tumor size, palpability, movability, lesion location and histopathology were analyzed. Results The average Emax, Emean, Esd and Eratio of the longest diameter were significantly higher than orthogonal diameter (P

We aimed to investigate the diagnostic performance of shear wave elastography (SWE) combined with conventional ultrasonography (US) for differentiating between benign and malignant thyroid nodules of different sizes.A total of 445 thyroid nodules from 445 patients were divided into 3 groups based on diameter (group 1, ≤ 10 mm; group 2, 10-20 mm; and group 3, 20 mm). The mean elasticity index of the whole lesion was automatically calculated, and the threshold for differentiation between benign and malignant nodules was constructed by a receiver operating characteristic curve analysis. Diagnostic performances of conventional US and SWE were compared by using pathologic results as reference standards.The mean elasticity was significantly higher in malignant versus benign nodules for all size groups. The differences in mean elasticity in the size groups were not statistically significant for malignant or benign nodules. The specificity of US combined with SWE for group 1 was significantly higher than that for groups 2 and 3 (77.8% versus 62.9% and 53.3%; P .05), and compared with group 1, the sensitivity was significantly higher for groups 2 and 3 (92.4% and 94.3% versus 80.7%; P .05). When SWE was added, the specificity increased and the sensitivity and diagnostic accuracy decreased for group 1, and the sensitivity increased and the specificity decreased for groups 2 and 3; however, the differences were not significant.Combined with SWE, US yielded higher specificity for nodules of 10 mm and smaller and higher sensitivity for nodules larger than 10 mm.

In current clinical practice, invasive ductal carcinoma is always screened using medical imaging techniques and diagnosed using immunohistochemistry. Recent studies have illustrated that radiomics approaches provide a comprehensive characterization of entire tumors and can reveal predictive or prognostic associations between the images and medical outcomes. To better reveal the underlying biology, an improved understanding between objective image features and biologic characteristics is urgently required.A total of 215 patients with definite histologic results were enrolled in our study. The tumors were automatically segmented using our phase-based active contour model. The high-throughput radiomics features were designed and extracted using a breast imaging reporting and data system and further selected using Student's t test, interfeature coefficients and a lasso regression model. The support vector machine classifier with threefold cross-validation was used to evaluate the relationship.The radiomics approach demonstrated a strong correlation between receptor status and subtypes (P .05; area under the curve, 0.760). The appearance of hormone receptor-positive cancer and human epidermal growth factor receptor 2-negative cancer on ultrasound scans differs from that of triple-negative cancer.Our approach could assist clinicians with the accurate prediction of prognosis using ultrasound findings, allowing for early medical management and treatment.

Background: Papillary thyroid carcinoma is a type of indolent tumor with a dramatically increasing incidence rate and stably high survival rate. Reducing the overdiagnosis and overtreatment of papillary thyroid carcinoma is clinically emergent and important. A radiomics model is proposed in this article to predict lymph node metastasis, the most important risk factor of papillary thyroid carcinoma, based on noninvasive routine preoperative ultrasound images. Methods: Four hundred fifty ultrasound manually segmented images of patients with papillary thyroid carcinoma with lymph node status obtained from pathology report were enrolled in our retrospective study. A radiomics evaluation of 614 high-throughput features were calculated, including size, shape, margin, boundary, orientation, position, echo pattern, posterior acoustic pattern, and calcification features. Then, combined feature selection strategy was used to select features with the greatest ability to discriminate lymph node status. A support vector machine classifier was employed to build and validate the prediction model. Another independent testing cohort was used to further evaluate the performance of the radiomics model. Results: Among 614 radiomics features, 50 selected features most reflecting echo pattern, posterior acoustic pattern, and calcification showed the superior lymph node status distinguishable performance with area under the receiver operating characteristic curve of 0.753, 0.740, and 0.743 separately when using each type of features predicting the lymph node status. The results of model based on all 50 final features predicting the lymph node status shown an area under the receiver operating characteristic curve of 0.782, and accuracy of 0.712. In the independent testing cohort, the proposed approach showed similar results, with area under the receiver operating characteristic curve of 0.727 and accuracy of 0.710. Conclusion: Papillary thyroid carcinoma with lymph node metastasis usually shows a complex echo pattern, posterior region homogeneity, and macrocalcification or multiple calcification. The radiomics model proposed in this article is a promising method for assessing the risk of papillary thyroid carcinoma metastasis noninvasively.

OBJECTIVES To analyze the value of acoustic radiation force impulse imaging quantification in characterization of breast lesions and to analyze the stiffness of glandular and subcutaneous fatty tissue in benign and malignant lesions. METHODS A total of 175 breast lesions (67 malignant and 108 benign) in 173 women were studied. With acoustic radiation force impulse imaging, shear wave velocity (SWV), which can reflect the stiffness of tissue, was measured within the lesion (internal SWV [SWVi]), in the boundary zone (boundary SWV [SWVb]), in normal-appearing glandular tissue (glandular SWV [SWVg]), and in subcutaneous fatty tissue (fatty SWV [SWVf]). All lesions underwent core needle biopsy or surgical excision. Differences among the SWV types in malignant and benign lesions were evaluated. We also assessed how different lesion types affected the SWV types. Receiver operating characteristic analysis was performed to assess the sensitivity and specificity of the SWV types in differentiating malignant from benign lesions. RESULTS Internal SWV was significantly higher than SWVb, SWVg, and SWVf; SWVb was significantly higher than SWVg and SWVf; and SWVg was significantly higher than SWVf in both malignant and benign groups (P < .05). All 4 SWV types were significantly higher in the malignant group than the benign group (P < .05). In the malignant group, grade 3 invasive ductal carcinoma had the highest SWVi. The sensitivity and specificity for differentiating malignant lesions were 55.2% and 95.8% for SWVi, 85.1% and 53.3% for SWVb, 68.2% and 69.6% for SWVg, and 67.2% and 66.0% for SWVf. CONCLUSIONS Acoustic radiation force impulse imaging has potential to characterize breast lesions, both internally and within the boundary zone, and to reflect changes in stiffness within surrounding glandular and subcutaneous fatty tissues caused by malignant tumors.

Aim to discuss whether the contrast enhanced ultrasound (CEUS) can effectively monitor the efficacy on neoadjuvant chemotherapy of breast cancer or not by analyzing the indicators on chemotherapy CEUS and breast cancer tumor biology, especially tumor microcirculation indicator on animal mode. Human breast cancer cell lines MCF-7 are planted under the skins of nude mice. By simulating clinical neoadjuvant chemotherapy regimen periodically inject CMF (cyclophosphamide, methotrexate and fluorouracil) into the experimental group, and normal saline into the control group. Then detect the data from CEUS and record the parameters: maximum intensity (IMAX), rise time (RT), time to peak (TTP) and mean transit time (mTT). Execute animal after CEUS, obtain tumor biological indicator and record parameters: micro vessel density (MVD), vascular endothelial growth factor receptors 1/2/3/4 (VEGFR-1/2/3/4) and tumor cells. In the aspect of tumor biological indicator, the experimental group after the first drug delivery: inter- and intra-group comparisons of VEGFR-1/4drop significantly. The experimental group after the second drug delivery: inter- and intra-group comparisons of MVD, VEGFR-1/3/4drop significantly. In the aspect of parameters on tumor CEUS, the experimental group after the first drug delivery: inter- and intra-group comparisons of IMAX drop significantly. The experimental group after the second drug delivery: inter- and intra-group comparisons of IMAX decrease steeply; while inter-and intra-group comparisons of TTP rise significantly. There are great changes about the intra-group comparisons of the number of tumor cells before and after the experiment. In the process of chemotherapy, it maintains the consistency of the changes of CEUS parameters IMAX and TTP, tumor microcirculation indicators MVD and VEGFR-1/3/4 and tumor cells. So CEUS has a potential to make an early prediction on the efficacy of neoadjuvant chemotherapy.

// Lu Cao 1,4,5 , Gang Cai 1,4,5 , Cai Chang 2,5 , Zhao-Zhi Yang 1,5 , Yan Feng 1,5 , Xiao-Li Yu 1,5 , Jin-Li Ma 1,5 , Jiong Wu 3,5 , Xiao-Mao Guo 1,5 and Jia-Yi Chen 1,4,5 1 Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Shanghai, China 2 Department of Ultrasound, Fudan University Shanghai Cancer Center, Shanghai, China 3 Department of Breast Surgery, Fudan University Shanghai Cancer Center, Shanghai, China 4 Department of Radiation Oncology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China 5 Shanghai Medical College, Fudan University, Shanghai, China Correspondence to: Jia-Yi Chen, email: // Keywords : breast cancer, cardiotoxicity, radiotherapy, trastuzumab, concurrent treatment Received : May 24, 2015 Accepted : September 23, 2015 Published : October 09, 2015 Abstract Purpose: To evaluate the influence of concurrent trastuzumab on the cardiotoxicity in patients receiving left-sided adjuvant radiotherapy. Materials and Methods: Medical records of stage I-III left-sided breast cancer patients, including 64 receiving concurrent trastuzumab with radiotherapy and 73 receiving radiotherapy alone were retrospectively reviewed. All of the patients had normal LVEF after adjuvant chemotherapy. Information of doses volume to cardiac structures was collected. Cardiac events were assessed according to CTC 2.0. Results: Median follow-up of LVEF and clinical assessment of cardiac function from the initiation of radiotherapy was 6.7 months (range 3–60.9) and 26 months (range 6.4–60.9), respectively. Grade 1 LVEF dysfunction occurred in 5 (7.8%) and 3 (4.1%) patients of the concurrent-trastuzumab and radiotherapy alone cohort, respectively. Trastuzumab was the only significant factor influencing absolute LVEF decrease in univariate analysis. In multivariate analysis of concurrent-trastuzumab cohort, IMC radiotherapy and start trastuzumab during radiotherapy were independent risk factors. For concurrent cohort, mean heart dose, as well as D 10 -D 30 , D 50 -D 55 , V 5 -V 20 of the heart and D 30 -D 45 , D 65 -D 75 , V 6 -V 15 of the LV were significantly higher in patients developing LVEF dysfunction. Conclusions: Concurrent trastuzumab and left-sided radiotherapy is well tolerated in terms of cardiotoxicity in patients with normal baseline cardiac function after adjuvant chemotherapy. However, increases in mean dose and low–dose volume of cardiac structures are associated with a higher risk of acute LVEF dysfunction.

Background. Left ventricular ejection fraction (LVEF) is used routinely to monitor cardiac dysfunction associated with breast cancer treatment. In this study the prevalence of early left ventricular diastolic dysfunction (LVDD) and its relationship to the dose-volume of the heart irradiated were evaluated in HER2-positive breast cancer patients undergoing concurrent trastuzumab and adjuvant radiotherapy (RT). Materials and Methods. Data from 40 breast cancer patients treated with concurrent trastuzumab and left-sided adjuvant RT between September 2011 and October 2012 were collected prospectively. For comparison, 32 patients treated with concurrent trastuzumab and right-sided adjuvant RT and 71 patients treated with left-sided RT alone were collected retrospectively. Echocardiography was obtained before RT, immediately following RT, and 3 and 6 months after RT. Doses to the heart and left ventricle (LV) were quantified. Results. Prior to RT with concurrent trastuzumab, 11 of 29 (left) and 8 of 25 (right) patients with normal baseline left ventricular diastolic function (LVDF) developed LVDD. In patients receiving left-sided RT alone, 12 of 61 patients with normal baseline LVDF developed LVDD. Dmean, D15–D40, D60–D70, and V3–V10 of the LV were significantly higher in patients who developed LVDD after concurrent trastuzumab and left-sided RT. In contrast, only two patients developed grade 1 LVEF decrease after both concurrent treatment and left-sided RT alone. Conclusion. Changes in LVDF compared with LVEF are more sensitive for early detection of cardiotoxicity. The dose-volume of the heart contributes significantly to the risk of LVDD in patients with left-sided breast cancer treated concurrently with trastuzumab. Implications for Practice: Abnormalities in diastolic function are more sensitive than changes in the left ventricular ejection fraction for detecting acute cardiotoxicity and are related to the dose-volume of the heart irradiated in patients with left-sided breast cancer receiving radiotherapy concurrently with trastuzumab. This result highlights the importance of decreasing the dose-volume of heart irradiated as a protective strategy in the treatment setting of concurrent trastuzumab and radiotherapy. Diastolic dysfunction may serve as a more sensitive tool for the early detection of cardiac damage and should be incorporated as a routine parameter in the functional monitoring of cardiotoxicity.

To analyze the diagnostic performance of shear wave elastography (SWE) in differentiating between benign and malignant breast lesions, with special emphasis on the value of the "stiff rim" sign, as compared with conventional ultrasonography (US).For this ethics committee-approved retrospective study, all patients provided verbal informed consent for the analysis of their imaging data. A total of 193 consecutive women (age range, 18-82 years; mean age, 46 years) with 193 breast lesions (56 malignant, 137 benign) were included. The stiff rim sign, a qualitative SWE feature, was evaluated at the display setting of less than 180 kPa and at 180 kPa. The quantitative SWE features were assessed. Sensitivity, specificity, the area under the receiver operating characteristic curve (Az), and positive and negative likelihood ratios were calculated for conventional US features, for SWE features, and for combined conventional US and SWE features.Among all qualitative and quantitative SWE features, the stiff rim sign at the display setting (180 kPa) showed the highest Az (0.918; 95% confidence interval [CI]: 0.870, 0.953), which was comparable to that for conventional US (0.891; 95% CI: 0.838, 0.931) (P = .40). The combination of the stiff rim sign at less than 180 kPa and conventional US features, with a positive likelihood ratio of 12.23 (95% CI: 11.5, 13.0) and a negative likelihood ratio of 0.02 (95% CI: 0.003, 0.1), yielded the highest Az (0.982; 95% CI: 0.951, 0.995)-higher than those for conventional US only and for any single SWE feature alone (P.001 for all)-and yielded higher sensitivity (98.2% [55 of 56]; 95% CI: 90.4%, 100.0%) and similar specificity (92.0% [126 of 137]; 95% CI: 86.1%, 95.9%) compared with conventional US (P.001 and P = .58, respectively). With use of this combination, the rate of recommendation of benign lesions for interventional procedures would decrease from 56.9% (78 of 137) (on the basis of conventional US features) to 15.3% (21 of 137), and 100.0% (56 of 56) (95% CI: 93.6%, 100.0%) malignant lesions would be correctly selected for biopsy.Adding SWE features, especially the stiff rim sign at the display setting (180 kPa), to conventional US has the potential to improve the differentiation of breast lesions.