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

PURPOSE: 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%). CONCLUSION: 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.

Primary MALT lymphoma of the breast: pathological and radiological characteristics.

PURPOSE: Primary Mucosa-associated lymphoid tissue (MALT) lymphoma is a rare diagnosis in the breast, and clinical diagnosis based on radiological features is often challenging. This study aimed to evaluate the clinicopathological, and radiological characteristics of the patients diagnosed with primary breast MALT lymphoma. METHODS: This study examined 18 cases of primary MALT lymphoma of the breast diagnosed at a single tertiary center between January 2002 to December 2020. Medical charts, radiological imaging and original pathology slides were reviewed for each case. RESULTS: All cases were female (gender assigned at birth) and presented with a palpable mass or an incidental imaging finding. Imaging presentation ranged from mammographic asymmetries, circumscribed masses, and ultrasound masses lacking suspicious features. Seventeen cases were biopsied under ultrasound; one received a diagnostic excision biopsy. Microscopic examination of the breast specimens demonstrated atypical small lymphocyte infiltration with plasmacytoid differentiation and rare lymphoepithelial lesions. Immunohistochemistry was performed in all cases and established the diagnosis. Most patients were treated with radiotherapy, and only three were treated with chemotherapy. The median follow-up period was 4 years and 7.5 months, and all patients were alive at the last follow-up. CONCLUSION: Primary MALT breast lymphomas are usually indolent and non-systemic, and local radiotherapy may effectively alleviate local symptoms. Radiological findings show overlap with benign morphological features, which can delay the diagnosis of this unusual etiology. Although further studies involving a larger cohort could help establish the clinical and radiological characteristics of primary breast MALT lymphomas, pathology remains the primary method of diagnosis. TRIAL REGISTRATION NUMBER: University Health Network Ethics Committee (CAPCR/UHN REB number 19-5844), retrospectively registered.

Problem Solving MRI to Reduce False-Positive Biopsy Related to Breast US: Conductivity vs. DWI vs. Abbreviated Contrast-Enhanced MRI.

BACKGROUND: While breast ultrasound (US) is a useful tool for diagnosing breast masses, it can entail false-positive biopsy results because of some overlapping features between benign and malignant breast masses and subjective interpretation. PURPOSE: To evaluate the performance of conductivity imaging for reducing false-positive biopsy results related to breast US, as compared to diffusion-weighted imaging (DWI) and abbreviated MRI consisting of one pre- and one post-contrast T1-weighted imaging. STUDY TYPE: Prospective. SUBJECTS: Seventy-nine women (median age, 44 years) with 86 Breast Imaging Reporting and Data System (BI-RADS) category 4 masses as detected by breast US. FIELD STRENGTH/SEQUENCE: 3-T, T2-weighted turbo spin echo sequence, DWI, and abbreviated contrast-enhanced MRI (T1-weighted gradient echo sequence). ASSESSMENT: US-guided biopsy (reference standard) was obtained on the same day as MRI. The maximum and mean conductivity parameters from whole and single regions of interest (ROIs) were measured. Apparent diffusion coefficient (ADC) values were obtained from an area with the lowest signal within a lesion on the ADC map. The performance of conductivity, ADC, and abbreviated MRI for reducing false-positive biopsies was evaluated using the following criteria: lowest conductivity and highest ADC values among malignant breast lesions and BI-RADS categories 2 or 3 on abbreviated MRI. STATISTICAL TESTS: One conductivity parameter with the maximum area under the curve (AUC) from receiver operating characteristics was selected. A P-value <0.05 was considered statistically significant. RESULTS: US-guided biopsy revealed 65 benign lesions and 21 malignant lesions. The mean conductivity parameter of the single ROI method was selected (AUC = 0.74). Considering conductivity (≤0.10 S/m), ADC (≥1.60 × 10-3 mm2 /sec), and BI-RADS categories 2 or 3 reduced false-positive biopsies by 23% (15 of 65), 38% (25 of 65), and 43% (28 of 65), respectively, without missing malignant lesions. DATA CONCLUSION: Conductivity imaging may show lower performance than DWI and abbreviated MRI in reducing unnecessary biopsies. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 2.

Detection and diagnosis of automated breast ultrasound in patients with BI-RADS category 4 microcalcifications: a retrospective study.

BACKGROUND: Automated Breast Ultrasound (AB US) has shown good application value and prospects in breast disease screening and diagnosis. The aim of the study was to explore the ability of AB US to detect and diagnose mammographically Breast Imaging Reporting and Data System (BI-RADS) category 4 microcalcifications. METHODS: 575 pathologically confirmed mammographically BI-RADS category 4 microcalcifications from January 2017 to June 2021 were included. All patients also completed AB US examinations. Based on the final pathological results, analyzed and summarized the AB US image features, and compared the evaluation results with mammography, to explore the detection and diagnostic ability of AB US for these suspicious microcalcifications. RESULTS: 250 were finally confirmed as malignant and 325 were benign. Mammographic findings including microcalcifications morphology (61/80 with amorphous, coarse heterogeneous and fine pleomorphic, 13/14 with fine-linear or branching), calcification distribution (189/346 with grouped, 40/67 with linear and segmental), associated features (70/96 with asymmetric shadow), higher BI-RADS category with 4B (88/120) and 4 C (73/38) showed higher incidence in malignant lesions, and were the independent factors associated with malignant microcalcifications. 477 (477/575, 83.0%) microcalcifications were detected by AB US, including 223 malignant and 254 benign, with a significantly higher detection rate for malignant lesions (x2 = 12.20, P < 0.001). Logistic regression analysis showed microcalcifications with architectural distortion (odds ratio [OR] = 0.30, P = 0.014), with amorphous, coarse heterogeneous and fine pleomorphic morphology (OR = 3.15, P = 0.037), grouped (OR = 1.90, P = 0.017), liner and segmental distribution (OR = 8.93, P = 0.004) were the independent factors which could affect the detectability of AB US for microcalcifications. In AB US, malignant calcification was more frequent in a mass (104/154) or intraductal (20/32), and with ductal changes (30/41) or architectural distortion (58/68), especially with the both (12/12). BI-RADS category results also showed that AB US had higher sensitivity to malignant calcification than mammography (64.8% vs. 46.8%). CONCLUSIONS: AB US has good detectability for mammographically BI-RADS category 4 microcalcifications, especially for malignant lesions. Malignant calcification is more common in a mass and intraductal in AB US, and tend to associated with architectural distortion or duct changes. Also, AB US has higher sensitivity than mammography to malignant microcalcification, which is expected to become an effective supplementary examination method for breast microcalcifications, especially in dense breasts.

Application of an artificial intelligence-based system in the diagnosis of breast ultrasound images obtained using a smartphone.

BACKGROUND: Breast ultrasound (US) is useful for dense breasts, and the introduction of artificial intelligence (AI)-assisted diagnoses of breast US images should be considered. However, the implementation of AI-based technologies in clinical practice is problematic because of the costs of introducing such approaches to hospital information systems (HISs) and the security risk of connecting HIS to the Internet to access AI services. To solve these problems, we developed a system that applies AI to the analysis of breast US images captured using a smartphone. METHODS: Training data were prepared using 115 images of benign lesions and 201 images of malignant lesions acquired at the Division of Breast Surgery, Gifu University Hospital. YOLOv3 (object detection models) was used to detect lesions on US images. A graphical user interface (GUI) was developed to predict an AI server. A smartphone application was also developed for capturing US images displayed on the HIS monitor with its camera and displaying the prediction results received from the AI server. The sensitivity and specificity of the prediction performed on the AI server and via the smartphone were calculated using 60 images spared from the training. RESULTS: The established AI showed 100% sensitivity and 75% specificity for malignant lesions and took 0.2 s per prediction with the AI sever. Prediction using a smartphone required 2 s per prediction and showed 100% sensitivity and 97.5% specificity for malignant lesions. CONCLUSIONS: Good-quality predictions were obtained using the AI server. Moreover, the quality of the prediction via the smartphone was slightly better than that on the AI server, which can be safely and inexpensively introduced into HISs.

Automated Breast Ultrasound (ABUS)-based radiomics nomogram: an individualized tool for predicting axillary lymph node tumor burden in patients with early breast cancer.

OBJECTIVES: Preoperative evaluation of axillary lymph node (ALN) status is an essential part of deciding the appropriate treatment. According to ACOSOG Z0011 trials, the new goal of the ALN status evaluation is tumor burden (low burden, < 3 positive ALNs; high burden, ≥ 3 positive ALNs), instead of metastasis or non-metastasis. We aimed to develop a radiomics nomogram integrating clinicopathologic features, ABUS imaging features and radiomics features from ABUS for predicting ALN tumor burden in early breast cancer. METHODS: A total of 310 patients with breast cancer were enrolled. Radiomics score was generated from the ABUS images. Multivariate logistic regression analysis was used to develop the predicting model, we incorporated the radiomics score, ABUS imaging features and clinicopathologic features, and this was presented with a radiomics nomogram. Besides, we separately constructed an ABUS model to analyze the performance of ABUS imaging features in predicting ALN tumor burden. The performance of the models was assessed through discrimination, calibration curve, and decision curve. RESULTS: The radiomics score, which consisted of 13 selected features, showed moderate discriminative ability (AUC 0.794 and 0.789 in the training and test sets). The ABUS model, comprising diameter, hyperechoic halo, and retraction phenomenon, showed moderate predictive ability (AUC 0.772 and 0.736 in the training and test sets). The ABUS radiomics nomogram, integrating radiomics score with retraction phenomenon and US-reported ALN status, showed an accurate agreement between ALN tumor burden and pathological verification (AUC 0.876 and 0.851 in the training and test sets). The decision curves showed that ABUS radiomics nomogram was clinically useful and more excellent than US-reported ALN status by experienced radiologists. CONCLUSIONS: The ABUS radiomics nomogram, with non-invasive, individualized and precise assessment, may assist clinicians to determine the optimal treatment strategy and avoid overtreatment.

Optoacoustic Imaging With Decision Support for Differentiation of Benign and Malignant Breast Masses: A 15-Reader Retrospective Study.

BACKGROUND. Overlap in ultrasound features of benign and malignant breast masses yields high rates of false-positive interpretations and benign biopsy results. Optoacoustic imaging is an ultrasound-based functional imaging technique that can increase specificity. OBJECTIVE. The purpose of this study was to compare specificity at fixed sensitivity of ultrasound images alone and of fused ultrasound and optoacoustic images evaluated with machine learning-based decision support tool (DST) assistance. METHODS. This retrospective Reader-02 study included 480 patients (mean age, 49.9 years) with 480 breast masses (180 malignant, 300 benign) that had been classified as BI-RADS category 3-5 on the basis of conventional gray-scale ultrasound findings. The patients were selected by stratified random sampling from the earlier prospective 16-site Pioneer-01 study. For that study, masses were further evaluated by ultrasound alone followed by fused ultrasound and optoacoustic imaging between December 2012 and September 2015. For the current study, 15 readers independently reviewed the previously acquired images after training in optoacoustic imaging interpretation. Readers first assigned probability of malignancy (POM) on the basis of clinical history, mammographic findings, and conventional ultrasound findings. Readers then evaluated fused ultrasound and optoacoustic images, assigned scores for ultrasound and optoacoustic imaging features, and viewed a POM prediction score derived by a machine learning-based DST before issuing final POM. Individual and mean specificities at fixed sensitivity of 98% and partial AUC (pAUC) (95-100% sensitivity) were calculated. RESULTS. Averaged across all readers, specificity at fixed sensitivity of 98% was significantly higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone (47.2% vs 38.2%; p = .03). Across all readers, pAUC was higher (p < .001) for fused ultrasound and optoacoustic imaging with DST assistance (0.024 [95% CI, 0.023-0.026]) than for ultrasound alone (0.021 [95% CI, 0.019-0.022]). Better performance using fused ultrasound and optoacoustic imaging with DST assistance than using ultrasound alone was observed for 14 of 15 readers for specificity at fixed sensitivity and for 15 of 15 readers for pAUC. CONCLUSION. Fused ultrasound and optoacoustic imaging with DST assistance had significantly higher specificity at fixed sensitivity than did conventional ultrasound alone. CLINICAL IMPACT. Optoacoustic imaging, integrated with reader training and DST assistance, may help reduce the frequency of biopsy of benign breast masses.

Utility of MRI-Directed Contrast-Enhanced Mammography for Biopsy Planning for Suspicious MRI-Detected Breast Lesions.

BACKGROUND. Suspicious lesions detected on contrast-enhanced breast MRI often undergo targeted ultrasound evaluation to determine whether they are amenable to ultrasound-guided biopsy. OBJECTIVE. The purpose of this study is to assess the utility of MRI-directed contrast-enhanced mammography (CEM) performed for biopsy planning for suspicious MRI-detected breast lesions and to compare its use with that of MRI-directed ultrasound. METHODS. This retrospective study included 120 patients (median age, 50.3 years) who underwent MRI-directed CEM from September 2014 to July 2020 for biopsy planning for a total of 140 suspicious breast MRI lesions; 109 lesions were also evaluated by MRI-directed ultrasound at the same visit. The reference standard was histopathology or at least 2 years of imaging follow-up for benign lesions. Rates of detecting a correlate for the MRI lesion, among all lesions and among malignant lesions, were compared between MRI-directed CEM, MRI-directed ultrasound, and combined MRI-directed CEM and ultrasound (i.e., with the correlate detected on either modality), by use of the McNemar test. The frequencies with which imaging modalities were used for biopsy guidance after MRI-directed imaging were determined. RESULTS. Twenty-three of 109 lesions were malignant. The lesion detection rate was higher for MRI-directed CEM than for MRI-directed ultrasound (69.7% [76/109] vs 45.9% [50/109]; p < .001) and higher for combined MRI-directed CEM and ultrasound (77.1% [84/109]) than for either MRI-directed CEM (p = .008) or MRI-directed ultrasound (p < .001). The rate of detection of malignant lesions was not significantly different between MRI-directed CEM and MRI-directed ultrasound (95.7% [22/23] vs 78.3% [18/23]; p = .13). A total of 31.2% (34/109) of lesions were seen on MRI-directed CEM only, and 7.3% (8/109) were seen on MRI-directed ultrasound only. A total of 17.4% (4/23) of malignant lesions were seen on MRI-directed CEM only, and none were seen on MRI-directed ultrasound only. Among lesions recommended for biopsy, stereotactic- or tomosynthesis-guided biopsy was recommended for 25.2% (26/103), ultrasound-guided biopsy for 35.9% (37/103), and MRI-guided biopsy for 38.8% (40/103). CONCLUSION. MRI-directed CEM detects a higher fraction of suspicious MRI lesions than does MRI-directed ultrasound. Combined MRI-directed CEM and ultrasound detects a higher fraction than either method does individually. CLINICAL IMPACT. MRI-directed CEM may be a useful alternate or complementary tool to MRI-directed ultrasound in biopsy planning for suspicious MRI lesions, facilitating the use of biopsy guidance methods other than MRI guidance.

Twinkle artifact in sonographic breast clip visualization.

The 'twinkle' or 'twinkling' artifact represents a phenomenon observed using color Doppler ultrasound that leads to a rapid alternation of color in and immediately behind an echogenic and highly reflective object. It occurs during sonographic examination of kidney stones, and has been also described in clips used for marking breast and axillary lesions.

Current trends in diagnostic and therapeutic management of the axilla in breast cancer patients receiving neoadjuvant therapy: results of the German-wide NOGGO MONITOR 24 survey.

PURPOSE: In the last 2 decades, the optimal management of the axilla in breast cancer patients receiving neoadjuvant chemotherapy (NACT) has been one of the most frequently discussed topics. Little is known about the attitudes of surgeons/radiologists towards new developments such as targeted axillary dissection. Therefore, the NOGGO conducted a survey to evaluate the current approach to axillary management. METHODS: A standardized digital questionnaire was sent out to > 200 departments in Germany between 7/2021 and 5/2022. The survey was supported by EUBREAST. RESULTS: In total, 116 physicians completed the survey. In cN0 patients scheduled to receive NACT, 89% of respondents recommended sentinel lymph node biopsy (SLNB) after NACT. In case of ypN1mi(sn), 44% advised no further therapy, while 31% proposed ALND and 25% axillary irradiation. 64% of respondents recommended a minimally invasive axillary biopsy to cN + patients. TAD was used at the departments of 82% of respondents and was offered to all cN + patients converting to ycN0 by 57% and only to selected patients, usually based on the number of suspicious nodes at time of presentation, by 43%. The most common marking technique was a clip/coil. 67% estimated that the detection rate of their marker was very good or good. CONCLUSION: This survey shows a heterogenous approach towards axillary management in the neoadjuvant setting in Germany. Most respondents follow current guidelines. Since only two-thirds of respondents experienced the detection rate of the marker used at their department as (very) good, future studies should focus on the comparative evaluation of different marking techniques.

Fully automatic tumor segmentation of breast ultrasound images with deep learning.

BACKGROUND: Breast ultrasound (BUS) imaging is one of the most prevalent approaches for the detection of breast cancers. Tumor segmentation of BUS images can facilitate doctors in localizing tumors and is a necessary step for computer-aided diagnosis systems. While the majority of clinical BUS scans are normal ones without tumors, segmentation approaches such as U-Net often predict mass regions for these images. Such false-positive problem becomes serious if a fully automatic artificial intelligence system is used for routine screening. METHODS: In this study, we proposed a novel model which is more suitable for routine BUS screening. The model contains a classification branch that determines whether the image is normal or with tumors, and a segmentation branch that outlines tumors. Two branches share the same encoder network. We also built a new dataset that contains 1600 BUS images from 625 patients for training and a testing dataset with 130 images from 120 patients for testing. The dataset is the largest one with pixel-wise masks manually segmented by experienced radiologists. Our code is available at https://github.com/szhangNJU/BUS_segmentation. RESULTS: The area under the receiver operating characteristic curve (AUC) for classifying images into normal/abnormal categories was 0.991. The dice similarity coefficient (DSC) for segmentation of mass regions was 0.898, better than the state-of-the-art models. Testing on an external dataset gave a similar performance, demonstrating a good transferability of our model. Moreover, we simulated the use of the model in actual clinic practice by processing videos recorded during BUS scans; the model gave very low false-positive predictions on normal images without sacrificing sensitivities for images with tumors. CONCLUSIONS: Our model achieved better segmentation performance than the state-of-the-art models and showed a good transferability on an external test set. The proposed deep learning architecture holds potential for use in fully automatic BUS health screening.

2S-BUSGAN: A Novel Generative Adversarial Network for Realistic Breast Ultrasound Image with Corresponding Tumor Contour Based on Small Datasets.

Deep learning (DL) models in breast ultrasound (BUS) image analysis face challenges with data imbalance and limited atypical tumor samples. Generative Adversarial Networks (GAN) address these challenges by providing efficient data augmentation for small datasets. However, current GAN approaches fail to capture the structural features of BUS and generated images lack structural legitimacy and are unrealistic. Furthermore, generated images require manual annotation for different downstream tasks before they can be used. Therefore, we propose a two-stage GAN framework, 2s-BUSGAN, for generating annotated BUS images. It consists of the Mask Generation Stage (MGS) and the Image Generation Stage (IGS), generating benign and malignant BUS images using corresponding tumor contours. Moreover, we employ a Feature-Matching Loss (FML) to enhance the quality of generated images and utilize a Differential Augmentation Module (DAM) to improve GAN performance on small datasets. We conduct experiments on two datasets, BUSI and Collected. Moreover, results indicate that the quality of generated images is improved compared with traditional GAN methods. Additionally, our generated images underwent evaluation by ultrasound experts, demonstrating the possibility of deceiving doctors. A comparative evaluation showed that our method also outperforms traditional GAN methods when applied to training segmentation and classification models. Our method achieved a classification accuracy of 69% and 85.7% on two datasets, respectively, which is about 3% and 2% higher than that of the traditional augmentation model. The segmentation model trained using the 2s-BUSGAN augmented datasets achieved DICE scores of 75% and 73% on the two datasets, respectively, which were higher than the traditional augmentation methods. Our research tackles imbalanced and limited BUS image data challenges. Our 2s-BUSGAN augmentation method holds potential for enhancing deep learning model performance in the field.

Analysis of background echotexture on automated breast ultrasound using BI-RADS and modified classification: Association with clinical features and mammographic density.

PURPOSE: To analyze BE on ABUS using BI-RADS and a modified classification in association with mammographic density and clinical features. METHODS: Menopausal status, parity, and family history of breast cancer were collected for 496 women who underwent ABUS and mammography. Three radiologists independently reviewed all ABUS BE and mammographic density. Statistical analyses including kappa statistics (κ) for interobserver agreement, Fisher's exact test, and univariate and multivariate multinomial logistic regression were performed. RESULTS: BE distribution between the two classifications and between each classification and mammographic density were associated (P < 0.001). BI-RADS homogeneous-fibroglandular (76.8%) and modified heterogeneous BE (71.3%, 75.7%, and 87.5% of mild, moderate, and marked heterogeneous background echotexture, respectively) tended to be dense. BE was correlated between BI-RADS homogeneous-fat and modified homogeneous background (95.1%) and between BI-RADS homogeneous-fibroglandular or heterogeneous (90.6%) and modified heterogeneous (86.9%) (P < 0.001). In multinomial logistic regression, age < 50 years was independently associated with heterogeneous BE (OR, 8.89, P = 0.003, in BI-RADS; OR, 3.74; P = 0.020 in modified classification). CONCLUSION: BI-RADS homogeneous-fat and modified homogeneous BE on ABUS was likely to be mammographically fatty. However, BI-RADS homogeneous-fibroglandular or heterogeneous BE might be classified as any modified BE. Younger age was independently associated with heterogeneous BE.

BUS-Net: Breast Tumour Detection Network for Ultrasound Images Using Bi-directional ConvLSTM and Dense Residual Connections.

Breast ultrasound (BUS) imaging has become one of the key imaging modalities for medical image diagnosis and prognosis. However, the manual process of lesion delineation from ultrasound images can incur various challenges concerning variable shape, size, intensity, curvature, or other medical priors of the lesion in the image. Therefore, computer-aided diagnostic (CADx) techniques incorporating deep learning-based neural networks are automatically used to segment the lesion from BUS images. This paper proposes an encoder-decoder-based architecture to recognize and accurately segment the lesion from two-dimensional BUS images. The architecture is utilized with the residual connection in both encoder and decoder paths; bi-directional ConvLSTM (BConvLSTM) units in the decoder extract the minute and detailed region of interest (ROI) information. BConvLSTM units and residual blocks help the network weigh ROI information more than the similar background region. Two public BUS image datasets, one with 163 images and the other with 42 images, are used. The proposed model is trained with the augmented images (ten forms) of dataset one (with 163 images), and test results are produced on the second dataset and the testing set of the first dataset-the segmentation performance yielding comparable results with the state-of-the-art segmentation methodologies. Similarly, the visual results show that the proposed approach for BUS image segmentation can accurately identify lesion contours and can potentially be applied for similar and larger datasets.

A Single-Center Audit of BI-RADS 3 Assessment Category Utilization in Mammography and Breast Ultrasound.

Purpose: To evaluate outcomes of breast lesions assessed at our institution as probably benign (Breast Imaging Reporting and Data System [BI-RADS] category 3) with an expected malignancy rate of less than or equal to 2 %. Methods: Average-risk women with a BI-RADS 3 assessment following mammographic and/or ultrasound evaluation at our institution between January 1 and December 31, 2017 were included. Cancer yield was calculated within 90 days and at 6-month intervals up to 36 months. Results: Among 517 women (median age, 52 years; range, 13-89 years) with a BI-RADS 3 assessment, 349 (67.5 %) underwent biopsy or completed follow-up imaging up to 36 months. One hundred and 68 (32.5 %) were lost to follow-up. Thirty of 349 (8.6 %) had their imaging upgraded and underwent biopsy, yielding six cancers (cancer yield, 6 of 349 women [1.7 %]). Among 569 lesions assessed as BI-RADS 3, 92 (16.2 %) were characterized by morphologic features other than those validated as probably benign in prospective clinical studies. Fifty three of 517 women (10.3 %) had follow-up beyond 24 months, and 24 (4.6 %) had follow-up beyond 36 months. Conclusion: Overall utilization of the BI-RADS 3 assessment category at our institution is appropriate with a 1.7 % cancer yield. However, the rate of loss to follow-up, percentage of non-validated findings assessed as probably benign, and redundancy in follow-up protocols are too high, and warrant intervention. A patient handout explaining the BI-RADS 3 assessment category and automatic scheduling of follow-up studies have been implemented at our center to address loss to follow-up.

An updated perspective on diagnostics and treatment of idiopathic granulomatous mastitis.

AIM: To present our experiences with the anti-inflammatory treatment of idiopathic granulomatous mastitis (IGM) and highlight the imaging and anamnestic specifics of its diagnosis. METHODS: Patients with acute inflammatory breast disease underwent ultrasound examination followed by a collection of anamnestic data, and histological analysis of the process was performed using core-cut bio psy, confirming IGM. Subsequently, anti-inflammatory treatment was administered, consisting of a combination of colchicine, vitamin E, and local compresses made from an infusion of Plantago lanceolata. We also recorded any additional treatments administered extra muros prior to histological analysis (such as antibio tics, surgical intervention, and time from onset of symptoms to confirmation of diagnosis). We analyzed the effect of the anti-inflammatory treatment administered, including the onset of improvement, adverse effects, recurrences, and duration of treatment required for symptom resolution. RESULTS: Between 2016 and 2022, we diagnosed and histologically confirmed IGM in 53 patients through bio psy. Of these, 45 (84.9%) underwent the anti-inflammatory treatment we proposed, while eight (15.1%) opted for a different form of therapy. Currently, 27 patients (60%) are without treatment and clinical manifestations. The average duration of treatment was 34 months, and improvement in the clinical condition was observed within 2-8 weeks (average of 3 months). Four patients (14.81%) reported dyspepsia as an adverse effect. Recurrence occurred in five patients (18.52%) after 1-36 months (average of 7 months). Patients (22, 81.48%) who completed the treatment are without difficulties for 3-70 months (average of 34 months). The remaining 18 patients (40%) are currently undergoing treatment, lasting 3-41 months (average of 19 months). CONCLUSION: Anti-inflammatory treatment with colchicine, along with supportive therapy (compresses made from an infusion of Plantago lanceolata and vitamin E), represents a promising trend in the therapy of IGM, with minimal adverse effects.

Establishment of a deep-learning system to diagnose BI-RADS4a or higher using breast ultrasound for clinical application.

Although the categorization of ultrasound using the Breast Imaging Reporting and Data System (BI-RADS) has become widespread worldwide, the problem of inter-observer variability remains. To maintain uniformity in diagnostic accuracy, we have developed a system in which artificial intelligence (AI) can distinguish whether a static image obtained using a breast ultrasound represents BI-RADS3 or lower or BI-RADS4a or higher to determine the medical management that should be performed on a patient whose breast ultrasound shows abnormalities. To establish and validate the AI system, a training dataset consisting of 4028 images containing 5014 lesions and a test dataset consisting of 3166 images containing 3656 lesions were collected and annotated. We selected a setting that maximized the area under the curve (AUC) and minimized the difference in sensitivity and specificity by adjusting the internal parameters of the AI system, achieving an AUC, sensitivity, and specificity of 0.95, 91.2%, and 90.7%, respectively. Furthermore, based on 30 images extracted from the test data, the diagnostic accuracy of 20 clinicians and the AI system was compared, and the AI system was found to be significantly superior to the clinicians (McNemar test, p < 0.001). Although deep-learning methods to categorize benign and malignant tumors using breast ultrasound have been extensively reported, our work represents the first attempt to establish an AI system to classify BI-RADS3 or lower and BI-RADS4a or higher successfully, providing important implications for clinical actions. These results suggest that the AI diagnostic system is sufficient to proceed to the next stage of clinical application.

Prediction model of axillary lymph node status using automated breast ultrasound (ABUS) and ki-67 status in early-stage breast cancer.

BACKGROUND: Automated breast ultrasound (ABUS) is a useful choice in breast disease diagnosis. The axillary lymph node (ALN) status is crucial for predicting the clinical classification and deciding on the treatment of early-stage breast cancer (EBC) and could be the primary indicator of locoregional recurrence. We aimed to establish a prediction model using ABUS features of primary breast cancer to predict ALN status. METHODS: A total of 469 lesions were divided into the axillary lymph node metastasis (ALNM) group and the no ALNM (NALNM) group. Univariate analysis and multivariate analysis were used to analyze the difference of clinical factors and ABUS features between the two groups, and a predictive model of ALNM was established. Pathological results were as the gold standard. RESULTS: Ki-67, maximum diameter (MD), posterior feature shadowing or enhancement and hyperechoic halo were significant risk factors for ALNM in multivariate logistic regression analysis (P < 0.05). The four risk factors were used to build the predictive model, and it achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.791 (95% CI: 0.751, 0.831). The accuracy, sensitivity and specificity of the prediction model were 72.5%, 69.1% and 75.26%. The positive predictive value (PPV) and negative predictive value (NPV) were 66.08% and 79.93%, respectively. Distance to skin, MD, margin, shape, internal echo pattern, orientation, posterior features, and hyperechoic halo showed significant differences between stage I and stage II (P < 0.001). CONCLUSION: ABUS features and Ki-67 can meaningfully predict ALNM in EBC and the prediction model may facilitate a more effective therapeutic schedule.

Multiple Bilateral Circumscribed Masses at Screening Breast Ultrasound: Outcomes of New or Enlarging Masses at Follow-Up.

BACKGROUND. Breast screening ultrasound (US) has limited specificity but is increasingly performed because of widening state and federal legislation regarding breast density. There is a need for evidence-based management guidelines. OBJECTIVE. The purpose of this study was to assess outcomes of new or enlarging oval circumscribed parallel masses in the setting of multiple bilateral circumscribed masses (MBCM) at sequential rounds of US screening. METHODS. In this retrospective study of women found to have MBCM on screening breast US without mammography abnormalities, longitudinal review was performed to identify development of any new or enlarging or changing masses. Outcomes were recorded using biopsy results or minimum of 12 months of follow-up as reference standards. Lesion characteristics, BI-RADS classification, breast density, patient age, demographics, and risk factors were reviewed. Statistical analysis included multivariable logistic regression analysis. RESULTS. There were 284 (2.4%) cases of MBCM in a total of 48,488 bilateral screening US examinations performed in 11,826 asymptomatic women between January 1, 2014, and July 31, 2019, that fit inclusion criteria. Of the 284 women (mean age, 46 years; range, 20-83 years), 150 (52.8%) subsequently developed 465 new, enlarging, and/or changing masses, 107 (23.0%) of which underwent biopsy. Of the 465 masses, 408 (87.7%) were oval circumscribed parallel masses and similar to other MBCM, and 57 (12.3%) were unique findings that were nonoval noncircumscribed masses. None of the new or enlarging oval circumscribed parallel masses were malignant. In total, the malignancy rate was 0% for women with MBCM with follow-up (median, 40.8 months; range, 12-75 months) and 0% for those that underwent biopsy (95% CI, 0-1.2%). Among women with concurrent MBCM and unique findings, four cancers were detected. Three were new irregular masses, and one previously oval mass changed in morphology to have new calcifications and an irregular border. A younger age was related to the likelihood of having enlarging masses (p < .001). CONCLUSION. In the setting of MBCM, new or enlarging oval circumscribed parallel masses are a common and benign event. Concurrent new irregular masses or previously oval masses that develop suspicious morphologic features should be carefully evaluated for malignancy. CLINICAL IMPACT. Breast radiologists who encounter new or enlarging oval circumscribed parallel masses with no suspicious morphologic change in the setting of MBCM can safely defer biopsies.

Imaging Surveillance Options for Individuals With a Personal History of Breast Cancer: AJR Expert Panel Narrative Review.

Annual surveillance mammography is recommended for breast cancer survivors on the basis of observational studies and meta-analyses showing reduced breast cancer mortality and improved quality of life. However, breast cancer survivors are at higher risk of subsequent breast cancer and have a fourfold increased risk of interval breast cancers compared with individuals without a personal history of breast cancer. Supplemental surveillance modalities offer increased cancer detection compared with mammography alone, but utilization is variable, and benefits must be balanced with possible harms of false-positive findings. In this review, we describe the current state of mammographic surveillance, summarize evidence for supplemental surveillance in breast cancer survivors, and explore a risk-based approach to selecting surveillance imaging strategies. Further research identifying predictors associated with increased risk of interval second breast cancers and development of validated risk prediction tools may help physicians and patients weigh the benefits and harms of surveillance breast imaging and decide on a personalized approach to surveillance for improved breast cancer outcomes.