Challenging Contrast-Enhanced Mammography-Guided Biopsies: Practical Approach Using Real-Time Multimodality Imaging and a Proposed Procedural Algorithm.

Contrast-enhanced mammography (CEM) is an emerging functional breast imaging technique that entails the acquisition of dual-energy digital mammographic images after IV administration of iodine-based contrast material. CEM-guided biopsy technology was introduced in 2019 and approved by the U.S. FDA in 2020. This technology's availability enables direct sampling of suspicious enhancement seen only on or predominantly on recombined CEM images and addresses a major obstacle to the clinical implementation of CEM technology. The literature describing clinical indications and procedural techniques of CEM-guided biopsy is scarce. This article describes our initial experience in performing challenging CEM-guided biopsies and proposes a step-by-step procedural algorithm designed to proactively address anticipated technical difficulties and thereby increase the likelihood of achieving successful targeting.

Imaging of Breast Cancers With Predilection for Nonmass Pattern of Growth: Invasive Lobular Carcinoma and DCIS-Does Imaging Capture It All?

OBJECTIVE. Invasive lobular carcinoma (ILC) and ductal carcinoma in situ (DCIS) are distinct histopathologic entities with several commonalities: both have subtle clinical and imaging presentation, have been linked with controversy regarding optimal imaging techniques and management, and exemplify the codependence of adequate imaging evaluation and optimal treatment strategies in breast care. CONCLUSION. We review molecular mechanisms and histopathologic patterns that define the biologic behavior of both ILC and DCIS and discuss how these mechanisms translate into distinct clinical and imaging presentations that affect the staging workup and patient management algorithm.

Sonographic features of benign and malignant axillary nodes post-neoadjuvant chemotherapy.

The aim was to determine whether sonographic features of metastatic axillary lymph nodes predict pathologic nodal status post-neoadjuvant chemotherapy (NCT) and help to tailor less invasive surgical management of the axilla. Patients with biopsy-proven cN1 primary breast malignancy who received NCT between January 2011 and December 2014 and had performed ultrasound were included in this study. Sonographic features of biopsy-proven clipped metastatic axillary nodes pre- and post-NCT were retrospectively reviewed by two independent readers. Changes in lymph node shape, fatty hilum status, cortical thickness, and cortical echogenicity were compared in patients with and without nodal pathologic complete response (pCR) using univariate and multivariate logistic regression models. Inter-reader variation was analyzed to determine the reproducibility of data. Of the 195 patients included in the study, 75 (45%) showed nodal pCR and 90 (55%) persistent metastatic disease post-NCT. pCR was significantly more likely in lymph nodes with isoechoic or hypoechoic cortical echogenicity post-NCT, (P = .02), conversion to normal cortical thickness (P = .0001), and oval shape (odds ratio = 0.17, P = .004), compared to lymph nodes with anechoic cortical echogenicity, persistent or diffuse cortical thickening, and irregular shape, respectively. The overall accuracy of sonographic nodal features in the prediction of pCR was 65% (95% CI: 58%-72%). The overall accuracy of sonographic features of biopsy-proven metastatic axillary lymph nodes post-NCT is not sufficiently high to predict pCR of axillary nodal status and thereby should not be solely used in guiding less invasive surgical approaches to the axilla.

Biomarkers and Imaging of Breast Cancer.

OBJECTIVE: The goals of this review are to provide background information on the definitions and applications of the general term "biomarker" and to highlight the specific roles of breast imaging biomarkers in research and clinical breast cancer care. A search was conducted of the main electronic biomedical databases (PubMed, Cochrane, Embase, MEDLINE [Ovid], Scopus, and Web of Science). The search was focused on review literature in general radiology and biomedical sciences and on reviews and primary research articles on biomarkers in breast imaging over the 15 years ending in June 2017. The keywords included "biomarker," "trial endpoints," "breast imaging," "breast cancer," "radiomics," and "precision medicine" in the titles and abstracts of the papers. CONCLUSION: Clinical breast care and breast cancer-related research rely on imaging biomarkers for decision support. In the era of precision medicine and big data, the practice of radiology is likely to change. A closer integration of breast imaging with related biomedical fields and the creation of large integrated and shareable databases of clinical, molecular, and imaging biomarkers should allow the field to continue guiding breast cancer care and research.

The Impact of Patient Demographics on the Selection of Breast Imaging Centers.

OBJECTIVE: Studies show that health care tailored to patient preferences results in significant improvements in physician performance, patient satisfaction, and health outcomes. Limited information in the literature exists on the factors driving patient preferences for establishing care at specific breast imaging centers. In this study, we identified factors that drive cohort preferences in the selection of a breast imaging center. MATERIALS AND METHODS: An 18-question survey was deployed in a large metropolitan area to gather information on patient demographics and preferences for breast imaging center location and radiologist training level. Cluster analysis and the K-means method were used to classify patients into groups on the basis of their answers about preference. Clusters were tested for significant differences by location, reason for visit, age, education, marital status, ethnicity, insurance, history of cancer, and income. RESULTS: A total of 1682 survey responses (18% of total patient visits) were obtained. Four distinct cohorts (comprising 876 patients) based on patient care preferences were identified: convenience optimizers (n = 109, 12.4%), ambivalent patients (n = 237, 27.1%), medical center seekers (n = 324, 37.0%), and expertise seekers (n = 206, 23.5%). Each cohort showed distinct preferences for imaging center location and radiologist training. Cohorts were differentiated on the basis of patient education level, ethnicity, and patient cancer history. Across the cohorts, there were no significant differences in age, marital status, insurance, income, and other demographic factors. CONCLUSION: Patient preferences for breast imaging care and location vary and are correlated with specific demographic characteristics. An understanding of these population characteristics can shape organizational strategies for improving patient-centered care and outcomes.

Deep Learning Models for Automated Assessment of Breast Density Using Multiple Mammographic Image Types.

Recently, convolutional neural network (CNN) models have been proposed to automate the assessment of breast density, breast cancer detection or risk stratification using single image modality. However, analysis of breast density using multiple mammographic types using clinical data has not been reported in the literature. In this study, we investigate pre-trained EfficientNetB0 deep learning (DL) models for automated assessment of breast density using multiple mammographic types with and without clinical information to improve reliability and versatility of reporting. 120,000 for-processing and for-presentation full-field digital mammograms (FFDM), digital breast tomosynthesis (DBT), and synthesized 2D images from 5032 women were retrospectively analyzed. Each participant underwent up to 3 screening examinations and completed a questionnaire at each screening encounter. Pre-trained EfficientNetB0 DL models with or without clinical history were optimized. The DL models were evaluated using BI-RADS (fatty, scattered fibroglandular densities, heterogeneously dense, or extremely dense) versus binary (non-dense or dense) density classification. Pre-trained EfficientNetB0 model performances were compared using inter-observer and commercial software (Volpara) variabilities. Results show that the average Fleiss' Kappa score between-observers ranged from 0.31-0.50 and 0.55-0.69 for the BI-RADS and binary classifications, respectively, showing higher uncertainty among experts. Volpara-observer agreement was 0.33 and 0.54 for BI-RADS and binary classifications, respectively, showing fair to moderate agreement. However, our proposed pre-trained EfficientNetB0 DL models-observer agreement was 0.61-0.66 and 0.70-0.75 for BI-RADS and binary classifications, respectively, showing moderate to substantial agreement. Overall results show that the best breast density estimation was achieved using for-presentation FFDM and DBT images without added clinical information. Pre-trained EfficientNetB0 model can automatically assess breast density from any images modality type, with the best results obtained from for-presentation FFDM and DBT, which are the most common image archived in clinical practice.

COVID-19 Vaccine-Related Axillary and Cervical Lymphadenopathy in Patients with Current or Prior Breast Cancer and Other Malignancies: Cross-Sectional Imaging Findings on MRI, CT, and PET-CT.

Breast radiologists are increasingly seeing patients with axillary adenopathy related to COVID-19 vaccination. Vaccination can cause levels I-III axillary as well as cervical lymphadenopathy. Appropriate management of vaccine-related adenopathy may vary depending on clinical context. In patients with current or past history of malignancy, vaccine-related adenopathy can be indistinguishable from nodal metastasis. This article presents imaging findings of oncology patients with adenopathy seen in the axilla or neck on cross-sectional imaging (breast MRI, CT, or PET-CT) after COVID-19 vaccination. Management approach and rationale is discussed, along with consideration on strategies to minimize false positives in vaccinated cancer patients. Time interval between vaccination and adenopathy seen on breast MRI, CT, or PET-CT is also reported.

Batch Reading and Interrupted Interpretation of Digital Screening Mammograms Without and With Tomosynthesis.

OBJECTIVE: To compare batch reading and interrupted interpretation for modern screening mammography. METHODS: We retrospectively reviewed digital mammograms without and with tomosynthesis that were originally interpreted with batch reading or interrupted interpretation between January 2015 and June 2017. The following performance metrics were compared: recall rate (per 100 examinations), cancer detection rate (per 1,000 examinations), and positive predictive values for recall and biopsy. RESULTS: In all, 9,832 digital mammograms were batch read, yielding a recall rate of 9.98%, cancer detection rate of 4.27, and positive predictive values for recall and biopsy of 4.40% and 35.5%, respectively. There were 49,496 digital mammograms that were read with interrupted interpretation, yielding a recall rate of 11.3%, cancer detection rate of 4.44, and positive predictive values for recall and biopsy of 3.92% and 30.1%, respectively. Of the digital mammograms with tomosynthesis, 7,075 were batch read, yielding a recall rate of 6.98%, cancer detection rate of 5.37, and positive predictive values for recall and biopsy of 7.69% and 38.0%, respectively. Of the digital mammograms with tomosynthesis, 24,380 were read with interrupted interpretation, yielding a recall rate of 8.30%, cancer detection rate of 5.41, and positive predictive values for recall and biopsy of 6.52% and 33.3%, respectively. For both digital mammograms without and with tomosynthesis, recall rates improved with batch reading compared with interrupted interpretation (P < .001), but no significant differences were seen for other metrics. DISCUSSION: Batch reading digital mammograms without and with tomosynthesis improves recall rates while maintaining cancer detection rates and positive predictive values compared with interrupted interpretation.

Does lateral arm technique decrease the rate of clip migration in stereotactic and tomosynthesis-guided biopsies?

BACKGROUND: Mammography-guided vacuum-assisted biopsies (MGVAB) can be done with stereotaxis or digital breast tomosynthesis guidance. Both methods can be performed with a conventional (CBA) or a lateral arm biopsy approach (LABA). Marker clip migration is relatively frequent in MGVAB (up to 44%), which in cases requiring surgery carries a risk of positive margins and re-excision. We aimed to compare the rates of clip migration and hematoma formation between the CBA and LABA techniques of prone MGVAB. Our HIPAA compliant retrospective study included all consecutive prone MGVAB performed in a single institution over a 20-month period. The LABA approach was used with DBT guidance; CBA utilized DBT or stereotactic guidance. The tissue sampling techniques were otherwise identical. RESULTS: After exclusion, 389 biopsies on 356 patients were analyzed. LABA was done in 97 (25%), and CBA in 292 (75%) cases. There was no statistical difference in clip migration rate with either 1 cm or 2 cm distance cut-off [15% for CBA and 10% for LABA for 1 cm threshold (p = 0.31); 5.8% or CBA and 3.1% or LABA for 2 cm threshold (p = 0.43)]. There was no difference in the rate of hematoma formation (57.5% in CDB and 50.5% in LABA, p = 0.24). The rates of technical failure were similar for both techniques (1.7% for CBA and 3% for LABA) with a combined failure rate of 1%. CONCLUSIONS: LABA and CBA had no statistical difference in clip migration or hematoma formation rates. Both techniques had similar success rates and may be helpful in different clinical situations.

A Deep Learning Approach to Re-create Raw Full-Field Digital Mammograms for Breast Density and Texture Analysis.

PURPOSE: To develop a computational approach to re-create rarely stored for-processing (raw) digital mammograms from routinely stored for-presentation (processed) mammograms. MATERIALS AND METHODS: In this retrospective study, pairs of raw and processed mammograms collected in 884 women (mean age, 57 years ± 10 [standard deviation]; 3713 mammograms) from October 5, 2017, to August 1, 2018, were examined. Mammograms were split 3088 for training and 625 for testing. A deep learning approach based on a U-Net convolutional network and kernel regression was developed to estimate the raw images. The estimated raw images were compared with the originals by four image error and similarity metrics, breast density calculations, and 29 widely used texture features. RESULTS: In the testing dataset, the estimated raw images had small normalized mean absolute error (0.022 ± 0.015), scaled mean absolute error (0.134 ± 0.078) and mean absolute percentage error (0.115 ± 0.059), and a high structural similarity index (0.986 ± 0.007) for the breast portion compared with the original raw images. The estimated and original raw images had a strong correlation in breast density percentage (Pearson r = 0.946) and a strong agreement in breast density grade (Cohen κ = 0.875). The estimated images had satisfactory correlations with the originals in 23 texture features (Pearson r ≥ 0.503 or Spearman ρ ≥ 0.705) and were well complemented by processed images for the other six features. CONCLUSION: This deep learning approach performed well in re-creating raw mammograms with strong agreement in four image evaluation metrics, breast density, and the majority of 29 widely used texture features.Keywords: Mammography, Breast, Supervised Learning, Convolutional Neural Network (CNN), Deep learning algorithms, Machine Learning AlgorithmsSee also the commentary by Chan in this issue.Supplemental material is available for this article.©RSNA, 2021.

Microcalcification is an important factor in the management of breast intraductal papillomas diagnosed on core biopsy.

The follow-up excision (FUE) results were analyzed from 370 cases diagnosed as intraductal papilloma on breast core needle biopsy (CNB) with no history of malignancy or other risk factors. Of these cases, 98.6% were rendered a Breast Imaging Reporting and Data System score of 4 on mammography before the CNB. Fifty-one cases (13.8%) were found to have microcalcifications on microscopic examination of CNB. A total of 7 (1.9%) of 370 cases were upgraded to invasive carcinoma, ductal carcinoma in situ, or pleomorphic lobular carcinoma in situ on FUE. Six of 51 (11.8%) cases with microcalcifications found on imaging and CNB were upgraded to ductal carcinoma in situ or invasive carcinoma, whereas only 1 (0.3%) of 319 cases without microcalcifications was upgraded to pleomorphic lobular carcinoma in situ (P = .003). Results of a multivariate analysis adjusted for age confirmed that microcalcifications was a risk factor for upgrading to cancer, independent of age. Our results indicate that surgical excision is required for intraductal papilloma diagnosed on CNB if microcalcifications are present. However, excision may not be required for those who have no microcalcifications on CNB and no other known risk factors.

Progressive pulmonary calcification in a child after orthotopic liver transplantation.

We present a case of progressive pulmonary calcification associated with prolonged respiratory insufficiency in a 2-year-old boy with a history of orthotopic liver transplantation. This case demonstrates the potentially progressive nature of pulmonary calcification and that it can present with respiratory insufficiency at a later period after transplantation than previously thought. We describe radiological findings and discuss established as well as plausible pathological mechanisms contributing to the development of calcifications in these patients.