Interoperator variability in the reproducibility of breast lesions found by handheld ultrasound (HHUS) can significantly interfere with clinical care. This study analyzed the features associated with breast mass position differences during HHUS. The ability of operators to reproduce the position of small masses and the time required to generate annotations with and without a computer-assisted scanning device (DEVICE) were also evaluated. This prospective study included 28 patients with 34 benign or probably benign small breast masses. Two operators generated manual and automated position annotations for each mass. The probe and body positions were systematically varied during scanning with the DEVICE, and the features describing mass movement were used in three logistic regression models trained to discriminate small from large breast mass displacements (cutoff: 10 mm). All models successfully discriminated small from large breast mass displacements (areas under the curve: 0.78 to 0.82). The interoperator localization precision was 6.6 ± 2.8 mm with DEVICE guidance and 19.9 ± 16.1 mm with manual annotations. Computer-assisted scanning reduced the time to annotate and reidentify a mass by 33 and 46 s on average, respectively. The results demonstrated that breast mass location reproducibility and exam efficiency improved by controlling operator actionable features with computer-assisted HHUS.

OBJECTIVE: To establish a reliable machine learning model to predict malignancy in breast lesions identified by ultrasound (US) and optimize the negative predictive value to minimize unnecessary biopsies.
METHODS: We included clinical and ultrasonographic attributes from 1526 breast lesions classified as BI-RADS 3, 4a, 4b, 4c, 5, and 6 that underwent US-guided breast biopsy in four institutions. We selected the most informative attributes to train nine machine learning models, ensemble models and models with tuned threshold to make inferences about the diagnosis of BI-RADS 4a and 4b lesions (validation dataset). We tested the performance of the final model with 403 new suspicious lesions.
RESULTS: The most informative attributes were shape, margin, orientation and size of the lesions, the resistance index of the internal vessel, the age of the patient and the presence of a palpable lump. The highest mean negative predictive value (NPV) was achieved with the K-Nearest Neighbors algorithm (97.9%). Making ensembles did not improve the performance. Tuning the threshold did improve the performance of the models and we chose the algorithm XGBoost with the tuned threshold as the final one. The tested performance of the final model was: NPV 98.1%, false negative 1.9%, positive predictive value 77.1%, false positive 22.9%. Applying this final model, we would have missed 2 of the 231 malignant lesions of the test dataset (0.8%).
CONCLUSIONS: Machine learning can help physicians predict malignancy in suspicious breast lesions identified by the US. Our final model would be able to avoid 60.4% of the biopsies in benign lesions missing less than 1% of the cancer cases.

Breast cancer is the most common cancer in women. Ultrasound is one of the most used techniques for diagnosis, but an expert in the field is necessary to interpret the test. Computer-aided diagnosis (CAD) systems aim to help physicians during this process. Experts use the Breast Imaging-Reporting and Data System (BI-RADS) to describe tumors according to several features (shape, margin, orientation...) and estimate their malignancy, with a common language. To aid in tumor diagnosis with BI-RADS explanations, this paper presents a deep neural network for tumor detection, description, and classification. An expert radiologist described with BI-RADS terms 749 nodules taken from public datasets. The YOLO detection algorithm is used to obtain Regions of Interest (ROIs), and then a model, based on a multi-class classification architecture, receives as input each ROI and outputs the BI-RADS descriptors, the BI-RADS classification (with 6 categories), and a Boolean classification of malignancy. Six hundred of the nodules were used for 10-fold cross-validation (CV) and 149 for testing. The accuracy of this model was compared with state-of-the-art CNNs for the same task. This model outperforms plain classifiers in the agreement with the expert (Cohen\'s kappa), with a mean over the descriptors of 0.58 in CV and 0.64 in testing, while the second best model yielded kappas of 0.55 and 0.59, respectively. Adding YOLO to the model significantly enhances the performance (0.16 in CV and 0.09 in testing). More importantly, training the model with BI-RADS descriptors enables the explainability of the Boolean malignancy classification without reducing accuracy.

OBJECTIVE: Evaluate surgical utilization of SCOUT reflectors placed at breast biopsy.
METHODS: Consent was waived for this retrospective IRB-approved, HIPAA-compliant study. Breast biopsy examinations that reported the term \"SCOUT\" between January 2021 and June 2022 were identified using an institutional search engine. Cases were included if a SCOUT reflector was placed at time of breast biopsy and excluded if lesion pathology was already known. Analysis was performed at the lesion level. A multivariate-regression analysis evaluated 6 variables with potential impact on SCOUT utilization.
RESULTS: One hundred twenty-one lesions in 112 patients met inclusion criteria. Biopsy yielded 93% (113/121) malignant, 3% (4/121) elevated risk, 2% (2/121) benign-discordant, and 2% (2/121) benign-concordant results. Two cases lost to follow-up were excluded. SCOUT reflectors were utilized for lumpectomy (58%, 69/119 lesions) and excisional biopsy (6%, 7/119 lesions). SCOUTs were not utilized due to mastectomy (23%, 27/119), subsequent wire localization (2%, 2/119), and nonsurgical cases (12%, 14/119). Reflector placement utilization was 52% higher for findings less than 3.5 cm in size (P <.001), 33% higher in patients without prior treated breast cancer (P = .012), and 19% higher in patients with no suspicious ipsilateral lymph node (P = .048).
CONCLUSIONS: SCOUT reflector placement at time of biopsy was utilized for surgery 64% (76/119) of the time, although most (98%, 119/121) biopsies were malignant, elevated risk, or benign-discordant. Factors increasing reflector utilization include smaller lesion size, no suspicious ipsilateral lymph node, and no prior treated breast cancer.

UNASSIGNED: To compare the medical image interpretation\'s time between the conventional and automated methods of breast ultrasound in patients with breast lesions. Secondarily, to evaluate the agreement between the two methods and interobservers.
UNASSIGNED: This is a cross-sectional study with prospective data collection. The agreement\'s degrees were established in relation to the breast lesions\'s ultrasound descriptors. To determine the accuracy of each method, a biopsy of suspicious lesions was performed, considering the histopathological result as the diagnostic gold standard.
UNASSIGNED: We evaluated 27 women. Conventional ultrasound used an average medical time of 10.77 minutes (± 2.55) greater than the average of 7.38 minutes (± 2.06) for automated ultrasound (p<0.001). The degrees of agreement between the methods ranged from 0.75 to 0.95 for researcher 1 and from 0.71 to 0.98 for researcher 2. Among the researchers, the degrees of agreement were between 0.63 and 1 for automated ultrasound and between 0.68 and 1 for conventional ultrasound. The area of the ROC curve for the conventional method was 0.67 (p=0.003) for researcher 1 and 0.72 (p<0.001) for researcher 2. The area of the ROC curve for the automated method was 0. 69 (p=0.001) for researcher 1 and 0.78 (p<0.001) for researcher 2.
UNASSIGNED: We observed less time devoted by the physician to automated ultrasound compared to conventional ultrasound, maintaining accuracy. There was substantial or strong to perfect interobserver agreement and substantial or strong to almost perfect agreement between the methods.

Several new ultrasound tools have been developed to further evaluate breast lesions detected on B-mode ultrasound. Strain elastography (SRE) was developed to assess the likelihood of malignancy of lesions based on their stiffness. This has been incorporated into the latest edition of the American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS) lexicon and atlas. However, no agreed cut-off stiffness values have been established to distinguish benign from malignant lesions making the translation into routine clinical practice difficult. Superb microvascular imaging (SMI) was developed to better evaluate the vascularity within sonographic lesions and assess their likelihood of malignancy. However, there is also no agreed cut-off value for vascular index (VI) to distinguish between benign and malignant lesions. MicroPure was developed to better visualize and evaluate calcifications seen on ultrasound. Its effective use in breast screening and evaluating the calcifications detected for likelihood of malignancy have not been established. This article describes the original intended uses of these applications and reviews the studies evaluating them, showing the varying success of the translation of these tools into routine clinical practice. Also described are some other uses of these tools for which they were not originally intended. This illustrates the importance of being perceptive to alternative uses of imaging tools in their translation from bench to bedside.

Background  Radial breast ultrasound scanning (r-US) and commonly used meander-like ultrasound scanning (m-US) have recently been shown to be equally sensitive and specific with regard to the detection of breast malignancies. As patient satisfaction has a strong influence on patient compliance and thus on the quality of health care, we compare here the two US scanning techniques with regard to patient comfort during breast ultrasound (BUS) and analyze whether the patient has a preference for either scanning technique. Materials and Methods  Symptomatic and asymptomatic women underwent both m-US and r-US scanning by two different examiners. Patient comfort and preference were assessed using a visual analog scale-based (VAS) questionnaire and were compared using a Mann-Whitney U test. Results  Analysis of 422 VAS-based questionnaires showed that perceived comfort with r-US (r-VAS 8 cm, IQR [5.3, 9.1]) was significantly higher compared to m-US (m-VAS 5.6 cm, IQR [5.2, 7.4]) (p < 0.001). 53.8% of patients had no preference, 44.3% of patients clearly preferred r-US, whereas only 1.9% of patients preferred m-US. Conclusion: Patients experience a higher level of comfort with r-US and favor r-US over m-US. As the diagnostic accuracy of r-US has been shown to be comparable to that of m-US and the time required for examination is shorter, a switch from m-US to r-US in routine clinical practice might be beneficial. R-US offers considerable potential to positively affect patient compliance but also to save examination time and thus costs.

Vascular tumors of the breast are rare, but benign hemangiomas are the most common type. Capillary hemangiomas are a subset of benign vascular tumors that involve smaller vessel sizes. They are difficult to diagnose with mammography and ultrasound, as they lack pathognomonic features and are frequently not seen. MRI is the most sensitive imaging tool. The lesions appear similar to angiosarcoma or ductal carcinoma in situ on imaging, which further complicates the diagnosis. A biopsy of the lesions is required for a definitive diagnosis. In this report, a 49-year-old female with newly diagnosed breast cancer is incidentally found to have a capillary hemangioma on staging breast MRI that was confirmed with a biopsy and excised along with the primary breast cancer with a partial mastectomy. The imaging findings of breast hemangioma on mammography, ultrasound, and MRI are also reviewed and described in this report.

UNASSIGNED: Accurate classification of breast nodules into benign and malignant types is critical for the successful treatment of breast cancer. Traditional methods rely on subjective interpretation, which can potentially lead to diagnostic errors. Artificial intelligence (AI)-based methods using the quantitative morphological analysis of ultrasound images have been explored for the automated and reliable classification of breast cancer. This study aimed to investigate the effectiveness of AI-based approaches for improving diagnostic accuracy and patient outcomes.
UNASSIGNED: In this study, a quantitative analysis approach was adopted, with a focus on five critical features for evaluation: degree of boundary regularity, clarity of boundaries, echo intensity, and uniformity of echoes. Furthermore, the classification results were assessed using five machine learning methods: logistic regression (LR), support vector machine (SVM), decision tree (DT), naive Bayes, and K-nearest neighbor (KNN). Based on these assessments, a multifeature combined prediction model was established.
UNASSIGNED: We evaluated the performance of our classification model by quantifying various features of the ultrasound images and using the area under the receiver operating characteristic (ROC) curve (AUC). The moment of inertia achieved an AUC value of 0.793, while the variance and mean of breast nodule areas achieved AUC values of 0.725 and 0.772, respectively. The convexity and concavity achieved AUC values of 0.988 and 0.987, respectively. Additionally, we conducted a joint analysis of multiple features after normalization, achieving a recall value of 0.98, which surpasses most medical evaluation indexes on the market. To ensure experimental rigor, we conducted cross-validation experiments, which yielded no significant differences among the classifiers under 5-, 8-, and 10-fold cross-validation (P>0.05).
UNASSIGNED: The quantitative analysis can accurately differentiate between benign and malignant breast nodules.

OBJECTIVE: Preventing positive margins is essential for ensuring favorable patient outcomes following breast-conserving surgery (BCS). Deep learning has the potential to enable this by automatically contouring the tumor and guiding resection in real time. However, evaluation of such models with respect to pathology outcomes is necessary for their successful translation into clinical practice.
METHODS: Sixteen deep learning models based on established architectures in the literature are trained on 7318 ultrasound images from 33 patients. Models are ranked by an expert based on their contours generated from images in our test set. Generated contours from each model are also analyzed using recorded cautery trajectories of five navigated BCS cases to predict margin status. Predicted margins are compared with pathology reports.
RESULTS: The best-performing model using both quantitative evaluation and our visual ranking framework achieved a mean Dice score of 0.959. Quantitative metrics are positively associated with expert visual rankings. However, the predictive value of generated contours was limited with a sensitivity of 0.750 and a specificity of 0.433 when tested against pathology reports.
CONCLUSIONS: We present a clinical evaluation of deep learning models trained for intraoperative tumor segmentation in breast-conserving surgery. We demonstrate that automatic contouring is limited in predicting pathology margins despite achieving high performance on quantitative metrics.