PPV of the Molecular Breast Imaging Lexicon.

BACKGROUND. Molecular breast imaging (MBI) is used for various breast imaging indications. An MBI lexicon has been developed, although the likelihood of malignancy of the lexicon descriptors has not been assessed to our knowledge. OBJECTIVE. The purpose of this article was to evaluate the PPV for malignancy of the MBI lexicon imaging descriptors. METHODS. This retrospective study included MBI examinations performed from August 1, 2005, through August 31, 2017, that were positive (BI-RADS analogous categories 0, 3, 4, 5, or 6) according to the clinical report and had an available reference standard. Examinations were performed using dual-detector cadmium zinc telluride MBI systems after injection of 99mTc sestamibi. Category 3 lesions had pathologic correlation, at least 2 years of imaging follow-up, or final resolution on follow-up imaging as category 1 or 2; category 4 and 5 lesions had pathologic correlation. MBI examinations were reviewed by one of two radiologists to assess lesions on the basis of the published MBI lexicon for type (mass vs nonmass uptake), distribution (if nonmass uptake), uptake intensity, and number of MBI views on which the lesion was seen. PPV for malignancy was summarized. RESULTS. The analysis included 643 lesions (479 benign, 164 malignant; 83 mass, 560 nonmass uptake) in 509 patients (median age, 56 years). PPV was 73.5% (61/83) for masses and 18.4% (103/560) for nonmass uptake. Among the nonmass uptake lesions, PPV was 36.2% (17/47) for segmental, 20.1% (77/384) for focal, 30.8% (4/13) for diffuse, and 4.3% (5/116) for regional or multiple regional distribution. PPV was 5.3% (5/94) for one view, 15.2% (32/210) for two views, 14.6% (13/89) for three views, and 45.4% (113/249) for four views showing the lesion. PPV was 14.0% (43/307) for mild, 22.4% (51/228) for moderate, and 64.8% (70/108) for marked uptake intensity. CONCLUSION. The MBI lexicon lesion descriptors are associated with likelihood of malignancy. PPV was higher for masses, lesions seen on multiple MBI views, and lesions with marked uptake intensity. Among nonmass uptake lesions, PPV was highest for those with segmental distribution. CLINICAL IMPACT. Insight into the likelihood of malignancy associated with the MBI lexicon descriptors can inform radiologists' interpretations and guide potential future incorporation of the MBI lexicon into the ACR BI-RADS Atlas.

Axillary Adenopathy after COVID-19 Vaccine: No Reason to Delay Screening Mammogram.

Online supplemental material is available for this article.

Non-BRCA Early-Onset Breast Cancer in Young Women.

The incidence of breast cancer in younger women is rising. Although early-onset breast cancer is highly associated with biologically aggressive tumors such as triple-negative and human epidermal growth factor 2 (HER2)-positive cancers, the more recent increase is disproportionately driven by an increase in the incidence of luminal cancer. In particular, the increase in de novo stage IV disease and the inherent age-based poorer survival rate among younger women with even early-stage luminal cancers suggest underlying distinct biologic characteristics that are not well understood. Further contributing to the higher number of early-onset breast cancers is pregnancy-associated breast cancer (PABC), which is attributed to persistent increases in maternal age over time. Although guidelines for screening of patients who carry a BRCA1 or BRCA2 gene mutation are well established, this population comprises only a fraction of those with early-onset breast cancer. A lack of screening in most young patients precludes timely diagnosis, underscoring the importance of early education and awareness. The disproportionate disease burden in young women of certain racial and ethnic groups, which is further exacerbated by socioeconomic disparity in health care, results in worse outcomes. An invited commentary by Monticciolo is available online. ©RSNA, 2022.

Imaging Features and Treatment Options for Breast Pseudoaneurysms After Biopsy: A Case-Based Pictorial Review.

Pseudoaneurysm (PSA) formation is a potential complication of breast biopsies. Ultrasound is the most common imaging modality used for evaluation and treatment of a PSA. Color Doppler images show a cystic structure with swirling flow inside in a "to-and-fro" pattern. Treatment options for PSA include observation, ultrasound-guided focused compression, thrombin injection, open surgical repair, and percutaneous embolization. The risks and benefits of these treatment options will be discussed in the following cases.

Clinical role of abbreviated and ultrafast MRI in breast imaging.

Current breast cancer screening relies on mammography, digital breast tomosynthesis and breast ultrasound. In select populations, breast MRI is also of great utility. However, multiple factors limit widespread use of breast MRI for screening. Efforts have been made to increase the availability of breast MRI for screening, in large part due to the increased cancer detection rate of breast MRI compared to mammography. Techniques include shortening standard breast MRI protocols with the potential for accommodating MRI screening in a higher number of patients. This review will explain the role of abbreviated breast MRI and ultrafast breast MRI in breast imaging, and detail how these approaches differ from standard dynamic contrast-enhanced breast MRI. In addition, limitations and advantages of these techniques will also be discussed.

Improving breast cancer diagnostics with deep learning for MRI.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has a high sensitivity in detecting breast cancer but often leads to unnecessary biopsies and patient workup. We used a deep learning (DL) system to improve the overall accuracy of breast cancer diagnosis and personalize management of patients undergoing DCE-MRI. On the internal test set (n = 3936 exams), our system achieved an area under the receiver operating characteristic curve (AUROC) of 0.92 (95% CI: 0.92 to 0.93). In a retrospective reader study, there was no statistically significant difference (P = 0.19) between five board-certified breast radiologists and the DL system (mean ΔAUROC, +0.04 in favor of the DL system). Radiologists' performance improved when their predictions were averaged with DL's predictions [mean ΔAUPRC (area under the precision-recall curve), +0.07]. We demonstrated the generalizability of the DL system using multiple datasets from Poland and the United States. An additional reader study on a Polish dataset showed that the DL system was as robust to distribution shift as radiologists. In subgroup analysis, we observed consistent results across different cancer subtypes and patient demographics. Using decision curve analysis, we showed that the DL system can reduce unnecessary biopsies in the range of clinically relevant risk thresholds. This would lead to avoiding biopsies yielding benign results in up to 20% of all patients with BI-RADS category 4 lesions. Last, we performed an error analysis, investigating situations where DL predictions were mostly incorrect. This exploratory work creates a foundation for deployment and prospective analysis of DL-based models for breast MRI.

Differences between human and machine perception in medical diagnosis.

Deep neural networks (DNNs) show promise in image-based medical diagnosis, but cannot be fully trusted since they can fail for reasons unrelated to underlying pathology. Humans are less likely to make such superficial mistakes, since they use features that are grounded on medical science. It is therefore important to know whether DNNs use different features than humans. Towards this end, we propose a framework for comparing human and machine perception in medical diagnosis. We frame the comparison in terms of perturbation robustness, and mitigate Simpson's paradox by performing a subgroup analysis. The framework is demonstrated with a case study in breast cancer screening, where we separately analyze microcalcifications and soft tissue lesions. While it is inconclusive whether humans and DNNs use different features to detect microcalcifications, we find that for soft tissue lesions, DNNs rely on high frequency components ignored by radiologists. Moreover, these features are located outside of the region of the images found most suspicious by radiologists. This difference between humans and machines was only visible through subgroup analysis, which highlights the importance of incorporating medical domain knowledge into the comparison.

Magnetic resonance imaging in the evaluation of pathologic nipple discharge: indications and imaging findings.

Pathologic nipple discharge (PND) is typically unilateral, spontaneous, involves a single duct, and is serous or bloody in appearance. In patients with PND, breast MRI can be helpful as an additional diagnostic tool when conventional imaging with mammogram and ultrasound are negative. MRI is able to detect the etiology of nipple discharge in 56-61% of cases when initial imaging with mammogram and ultrasound are negative. Advantages to using MRI in evaluation of PND include good visualization of the retroareolar breast and better evaluation of posterior lesions which may not be well evaluated on mammograms and galactograms. It is also less invasive compared to central duct excision. Papillomas and nipple adenomas are benign breast masses that can cause PND and are well visualized on MRI. Ductal ectasia, and infectious etiologies such as mastitis, abscess, and fistulas are additional benign causes of PND that are well evaluated with MRI. MRI is also excellent for evaluation of malignant causes of PND including Paget's disease, ductal carcinoma in-situ and invasive carcinoma. MRI's high negative predictive value of 87-98.2% is helpful in excluding malignant etiologies of PND.

MR Elastography of the Breast: Evolution of Technique, Case Examples, and Future Directions.

Recognizing that breast cancers present as firm, stiff lesions, the foundation of breast magnetic resonance elastography (MRE) is to combine tissue stiffness parameters with sensitive breast MR contrast-enhanced imaging. Breast MRE is a non-ionizing, cross-sectional MR imaging technique that provides for quantitative viscoelastic properties, including tissue stiffness, elasticity, and viscosity, of breast tissues. Currently, the technique continues to evolve as research surrounding the use of MRE in breast tissue is still developing. In the setting of a newly diagnosed cancer, associated desmoplasia, stiffening of the surrounding stroma, and necrosis are known to be prognostic factors that can add diagnostic information to patient treatment algorithms. In fact, mechanical properties of the tissue might also influence breast cancer risk. For these reasons, exploration of breast MRE has great clinical value. In this review, we will: (1) address the evolution of the various MRE techniques; (2) provide a brief overview of the current clinical studies in breast MRE with interspersed case examples; and (3) suggest directions for future research.

Screening Breast MRI Primer: Indications, Current Protocols, and Emerging Techniques.

Breast dynamic contrast-enhanced MRI (DCE-MRI) is the most sensitive imaging modality for the detection of breast cancer. Screening MRI is currently performed predominantly in patients at high risk for breast cancer, but it could be of benefit in patients at intermediate risk for breast cancer and patients with dense breasts. Decreasing scan time and image interpretation time could increase cost-effectiveness, making screening MRI accessible to a larger group of patients. Abbreviated breast MRI (Ab-MRI) reduces scan time by decreasing the number of sequences obtained, but as multiple delayed contrast enhanced sequences are not obtained, no kinetic information is available. Ultrafast techniques rapidly acquire multiple sequences during the first minute of gadolinium contrast injection and provide information about both lesion morphology and vascular kinetics. Diffusion-weighted imaging is a noncontrast MRI technique with the potential to detect mammographically occult cancers. This review article aims to discuss the current indications of breast MRI as a screening tool, examine the standard breast DCE-MRI technique, and explore alternate screening MRI protocols, including Ab-MRI, ultrafast MRI, and noncontrast diffusion-weighted MRI, which can decrease scan time and interpretation time.

Comparing deep learning-based automatic segmentation of breast masses to expert interobserver variability in ultrasound imaging.

Deep learning is a powerful tool that became practical in 2008, harnessing the power of Graphic Processing Unites, and has developed rapidly in image, video, and natural language processing. There are ongoing developments in the application of deep learning to medical data for a variety of tasks across multiple imaging modalities. The reliability and repeatability of deep learning techniques are of utmost importance if deep learning can be considered a tool for assisting experts, including physicians, radiologists, and sonographers. Owing to the high costs of labeling data, deep learning models are often evaluated against one expert, and it is unknown if any errors fall within a clinically acceptable range. Ultrasound is a commonly used imaging modality for breast cancer screening processes and for visually estimating risk using the Breast Imaging Reporting and Data System score. This process is highly dependent on the skills and experience of the sonographers and radiologists, thereby leading to interobserver variability and interpretation. For these reasons, we propose an interobserver reliability study comparing the performance of a current top-performing deep learning segmentation model against three experts who manually segmented suspicious breast lesions in clinical ultrasound (US) images. We pretrained the model using a US thyroid segmentation dataset with 455 patients and 50,993 images, and trained the model using a US breast segmentation dataset with 733 patients and 29,884 images. We found a mean Fleiss kappa value of 0.78 for the performance of three experts in breast mass segmentation compared to a mean Fleiss kappa value of 0.79 for the performance of experts and the optimized deep learning model.

Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams.

Though consistently shown to detect mammographically occult cancers, breast ultrasound has been noted to have high false-positive rates. In this work, we present an AI system that achieves radiologist-level accuracy in identifying breast cancer in ultrasound images. Developed on 288,767 exams, consisting of 5,442,907 B-mode and Color Doppler images, the AI achieves an area under the receiver operating characteristic curve (AUROC) of 0.976 on a test set consisting of 44,755 exams. In a retrospective reader study, the AI achieves a higher AUROC than the average of ten board-certified breast radiologists (AUROC: 0.962 AI, 0.924 ± 0.02 radiologists). With the help of the AI, radiologists decrease their false positive rates by 37.3% and reduce requested biopsies by 27.8%, while maintaining the same level of sensitivity. This highlights the potential of AI in improving the accuracy, consistency, and efficiency of breast ultrasound diagnosis.

Architectural Distortion on Digital Breast Tomosynthesis: Management Algorithm and Pathological Outcome.

Architectural distortion on digital breast tomosynthesis (DBT) can occur due to benign and malignant causes. With DBT, there is an increase in the detection of architectural distortion compared with 2D digital mammography, and the positive predictive value is high enough to justify tissue sampling when imaging findings are confirmed. Workup involves supplemental DBT views and ultrasound, with subsequent image-guided percutaneous biopsy using the modality on which it is best visualized. If architectural distortion is subtle and/or questionable on diagnostic imaging, MRI may be performed for problem solving, with subsequent biopsy of suspicious findings using MRI or DBT guidance, respectively. If no suspicious findings are noted on MRI, a six-month follow-up DBT may be performed. On pathology, malignant cases are noted in 6.8%-50.7% of the cases, most commonly due to invasive ductal carcinoma, followed by invasive lobular carcinoma. Radial scars are the most common benign cause, with stromal fibrosis and sclerosing adenosis being much less common. As there is an increase in the number of benign pathological outcomes for architectural distortion on DBT compared with 2D digital mammography, concordance should be based on the level of suspicion of imaging findings. As discordant cases have upgrade rates of up to 25%, surgical consultation is recommended for discordant radiologic-pathologic findings.

Correction to: Best MRI sequences for identifying axillary lymph node markers in patients with metastatic breast cancer: an inter-reader observational study.

An amendment to this paper has been published and can be accessed via the original article.

Best MRI sequences for identifying axillary lymph node markers in patients with metastatic breast cancer: an inter-reader observational study.

BACKGROUND: We assessed confidence in visualization of markers within metastatic axillary lymph nodes (LNs) on magnetic resonance imaging (MRI), which were placed post-ultrasound (US)-guided biopsy. METHODS: A retrospective review was performed on 55 MRI cases between May 2015 and October 2017. Twenty-two MRIs were performed before neoadjuvant therapy, and 33 MRIs were after its initiation. There were 34/55 HydroMARK®, 10/55 Tumark®, and 11/55 other marker types. Time interval between marker placement and MRI examination was 103 ± 81 days (mean ± standard deviation). Three readers with 1-30 years of experience independently assessed four axial sequences: unenhanced fat-suppressed three-dimensional T1-weighted spoiled gradient-recalled (SPGR), first contrast-enhanced fat-suppressed SPGR, T2-weighted water-only fast spin-echo (T2-WO), and T2-weighted fat-only fast-spin-echo (T2-FO). RESULTS: Markers were 5.2× more likely to be visualized on T2-WO than on unenhanced images (p = < 0.001), and 3.3× more likely to be visualized on contrast-enhanced than on unenhanced sequences (p = 0.009). HydroMARK markers demonstrated a 3× more likelihood of being visualized than Tumark (p = 0.003). Markers were 8.4× more likely to be visualized within morphologically abnormal LNs (p < 0.001). After 250 days post-placement, confidence in marker brightness of HydroMARK markers on T2-WO images was less than 50% (p < 0.001). Inter-rater agreement was excellent for T2-WO and contrast-enhanced SPGR, good for unenhanced SPGR, and poor for T2-FO images. CONCLUSION: T2-WO and contrast-enhanced images should be used for marker identification. HydroMARK was the best visualized marker. Markers were easier to identify when placed in abnormal LNs. The visibility of HydroMARK markers was reduced with time.

Harnessing the Power of Low-tech Collaborative Learning.

Medical education in the United States has undergone a paradigm shift from passive learning to more interactive student-centered teaching methods. Several digital tools and platforms have been developed to assist educators in creating a high-tech, interactive classroom. However, there are many low-tech ways to engage learners and improve retention by combining collaborative learning techniques and summary exercises. Collaborative learning is an educational approach that involves the coordinated engagement of two or more learners for the purpose of completing a task or solving a problem. Such methods use a student-centered active learning model to promote higher-order cognitive tasks through active engagement in course content. Additionally, summary exercises at the end of a learning activity promote reflection and retention of learned concepts while clarifying content that may have been confusing for the learner. The purpose of this article is to describe the methodology and tips for the implementation of low-tech collaborative learning methods and summary activities into trainee educational activities to create an engaging student-centered learning environment.

Imaging and Management of Internal Mammary Lymph Nodes.

Internal mammary lymph nodes (IMLNs) account for approximately 10%-40% of the lymphatic drainage of the breast. Internal mammary lymph nodes measuring up to 10 mm are commonly seen on high-risk screening breast MRI examinations in patients without breast cancer and are considered benign if no other suspicious findings are present. Benign IMLNs demonstrate a fatty hilum, lobular or oval shape, and circumscribed margins without evidence of central necrosis, cortical thickening, or loss of fatty hilum. In patients with breast cancer, IMLN involvement can alter clinical stage and treatment planning. The incidence of IMLN metastases detected on US, CT, MRI, and PET-CT ranges from 10%-16%, with MRI and PET-CT demonstrating the highest sensitivities. Although there are no well-defined imaging criteria in the eighth edition of the American Joint Committee on Cancer Staging Manual for Breast Cancer, a long-axis measurement of ≥ 5 mm is suggested as a guideline to differentiate benign versus malignant IMLNs in patients with newly diagnosed breast cancer. Abnormal morphology such as loss of fatty hilum, irregular shape, and rounded appearance (which can be quantified by a short-axis/long-axis length ratio greater than 0.5) also raises suspicion for IMLN metastases. MRI and PET-CT have good sensitivity and specificity for the detection of IMLN metastases, but fluorodeoxyglucose avidity can be seen in both benign conditions and metastatic disease. US is helpful for staging, and US-guided fine-needle aspiration can be performed in cases of suspected IMLN metastasis. Management of suspicious IMLNs identified on imaging is typically with chemotherapy and radiation, as surgical excision does not provide survival benefit and is performed only in rare cases.

Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening.

We present a deep convolutional neural network for breast cancer screening exam classification, trained, and evaluated on over 200000 exams (over 1000000 images). Our network achieves an AUC of 0.895 in predicting the presence of cancer in the breast, when tested on the screening population. We attribute the high accuracy to a few technical advances. 1) Our network's novel two-stage architecture and training procedure, which allows us to use a high-capacity patch-level network to learn from pixel-level labels alongside a network learning from macroscopic breast-level labels. 2) A custom ResNet-based network used as a building block of our model, whose balance of depth and width is optimized for high-resolution medical images. 3) Pretraining the network on screening BI-RADS classification, a related task with more noisy labels. 4) Combining multiple input views in an optimal way among a number of possible choices. To validate our model, we conducted a reader study with 14 readers, each reading 720 screening mammogram exams, and show that our model is as accurate as experienced radiologists when presented with the same data. We also show that a hybrid model, averaging the probability of malignancy predicted by a radiologist with a prediction of our neural network, is more accurate than either of the two separately. To further understand our results, we conduct a thorough analysis of our network's performance on different subpopulations of the screening population, the model's design, training procedure, errors, and properties of its internal representations. Our best models are publicly available at https://github.com/nyukat/breast_cancer_classifier.