Faculty Bibliography

Machine-learning techniques have led to remarkable advances in data extraction and analysis of medical imaging. Applications of machine learning to breast MRI continue to expand rapidly as increasingly accurate 3D breast and lesion segmentation allows the combination of radiologist-level interpretation (eg, BI-RADS lexicon), data from advanced multiparametric imaging techniques, and patient-level data such as genetic risk markers. Advances in breast MRI feature extraction have led to rapid dataset analysis, which offers promise in large pooled multiinstitutional data analysis. The object of this review is to provide an overview of machine-learning and deep-learning techniques for breast MRI, including supervised and unsupervised methods, anatomic breast segmentation, and lesion segmentation. Finally, it explores the role of machine learning, current limitations, and future applications to texture analysis, radiomics, and radiogenomics. Level of Evidence: 3 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2019.

Screening for breast cancer reduces breast cancer-related mortality and earlier detection facilitates less aggressive treatment. Unfortunately, current screening modalities are imperfect, suffering from limited sensitivity and high false-positive rates. Novel techniques in the field of breast imaging may soon play a role in breast cancer screening: digital breast tomosynthesis, contrast material-enhanced spectral mammography, US (automated three-dimensional breast US, transmission tomography, elastography, optoacoustic imaging), MRI (abbreviated and ultrafast, diffusion-weighted imaging), and molecular breast imaging. Artificial intelligence and radiomics have the potential to further improve screening strategies. Furthermore, nonimaging-based screening tests such as liquid biopsy and breathing tests may transform the screening landscape. © RSNA, 2020 Online supplemental material is available for this article.

Breast MRI offers high sensitivity for breast cancer detection, with preferential detection of high-grade invasive cancers when compared to mammography and ultrasound. Despite the clear benefits of breast MRI in cancer screening, its cost, patient tolerance, and low utilization remain key issues. Abbreviated breast MRI, in which only a select number of sequences and postcontrast imaging are acquired, exploits the high sensitivity of breast MRI while reducing table time and reading time to maximize availability, patient tolerance, and accessibility. Worldwide studies of varying patient populations have demonstrated that the comparable diagnostic accuracy of abbreviated breast MRI is comparable to a full diagnostic protocol, highlighting the emerging role of abbreviated MRI screening in patients with an intermediate and high lifetime risk of breast cancer. The purpose of this review is to summarize the background and current literature relating to abbreviated MRI, highlight various protocols utilized in current multicenter clinical trials, describe workflow and clinical implementation issues, and discuss the future of abbreviated protocols, including advanced MRI techniques.

Current guidelines recommend sentinel lymph node biopsy (SLNB) for patients undergoing mastectomy for a preoperative diagnosis of ductal carcinoma in situ (DCIS). We examined the factors associated with sentinel lymph node positivity for patients undergoing mastectomy for a diagnosis of DCIS on preoperative core biopsy (PCB). The Institutional Breast Cancer Database was queried for patients with PCB demonstrating pure DCIS followed by mastectomy and SLNB from 2010 to 2018. Patients were divided according to final pathology (DCIS or invasive cancer). Clinico-pathologic variables were analyzed using Pearson's chi-squared, Wilcoxon Rank-Sum and logistic regression. Of 3145 patients, 168(5%) had pure DCIS on PCB and underwent mastectomy with SLNB. On final mastectomy pathology, 120(71%) patients had DCIS with 0 positive sentinel lymph nodes (PSLNs) and 48(29%) patients had invasive carcinoma with 5(10%) cases of ≥1 PSLNs. Factors positively associated with upstaging to invasive cancer in univariate analysis included age (P = .0289), palpability (P < .0001), extent of disease on imaging (P = .0121), mass on preoperative imaging (P = .0003), multifocality (P = .0231) and multicentricity (P = .0395). In multivariate analysis, palpability (P = .0080), extent of disease on imaging (P = .0074) and mass on preoperative imaging (P = .0245) remained significant (Table 2). In a subset of patients undergoing mastectomy for DCIS with limited disease on preoperative evaluation, SLNB may be omitted as the risk of upstaging is low. However, patients who present with clinical findings of palpability, large extent of disease on imaging and mass on preoperative imaging have a meaningful risk of upstaging to invasive cancer, and SLNB remains important for management.

RATIONALE AND OBJECTIVES/OBJECTIVE:Pathologic complete response (pCR) in patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer after HER2-targeted therapy correlates increased disease-free survival and decreased mastectomy rates. The aim of this study was to explore tumor shrinkage patterns and initial tumor enhancement with pCR in HER2-positive breast cancer. MATERIALS AND METHODS/METHODS:This was an institutional review board-approved retrospective analysis of 51 HER2 positive breast cancer patients with breast MRI both pre- and post-HER2-targeted therapy. Initial enhancement ratio (IER, initial enhancement percentage over baseline at first postcontrast imaging), pattern of tumor shrinkage, and Dynamic contrast enhanced (DCE)-MRI imaging features were assessed. Wilcoxon rank, Spearman correlation, Fisher's exact, and Mann-Whitney tests were used to correlate MRI imaging features with pCR. IER reader agreement was evaluated by intraclass correlation. Binary logistic regression was used to evaluate multivariate associations with pCR. RESULTS:56.9% (29/51) of patients had pCR at surgery. Concentric tumor shrinkage pattern was associated with pCR (p = 0.001, Area under the curve (AUC) 0.778): accuracy 80.4%, specificity 96.6%, and sensitivity of 59.1%. There was no association with pCR and imaging response as defined by RECIST criteria (p = 0.169), pretreatment IER (Reader 1 (R1) p = 0.665, Reader 2 (R2) p = 0.766), or lesion size (p = 0.69). IER was associated with axillary metastases (R1 p = 0.016, R2 < 0.001) and ki-67 (R1 r = 0.52, p = 0.008, R2 r = -0.44, p = 0.028). CONCLUSION/CONCLUSIONS:The shrinkage pattern of HER2-positive tumors after targeted therapy may be associated with pCR. There was no association between IER and pCR. Future studies evaluating the correlation of shrinkage patterns to texture radiomics are of interest.

Axillary lymph node (LN) metastasis is the most important predictor of overall recurrence and survival in patients with breast cancer, and accurate assessment of axillary LN involvement is an essential component in staging breast cancer. Axillary management in patients with breast cancer has become much less invasive and individualized with the introduction of sentinel LN biopsy (SLNB). Emerging evidence indicates that axillary LN dissection may be avoided in selected patients with node-positive as well as node-negative cancer. Thus, assessment of nodal disease burden to guide multidisciplinary treatment decision making is now considered to be a critical role of axillary imaging and can be achieved with axillary US, MRI, and US-guided biopsy. For the node-positive patients treated with neoadjuvant chemotherapy, restaging of the axilla with US and MRI and targeted axillary dissection in addition to SLNB is highly recommended to minimize the false-negative rate of SLNB. Efforts continue to develop prediction models that incorporate imaging features to predict nodal disease burden and to select proper candidates for SLNB. As methods of axillary nodal evaluation evolve, breast radiologists and surgeons must work closely to maximize the potential role of imaging and to provide the most optimized treatment for patients.

We present a deep convolutional neural network for breast cancer screening exam classification, trained and evaluated on over 200,000 exams (over 1,000,000 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. (i) 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. (ii) 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. (iii) Pretraining the network on screening BI-RADS classification, a related task with more noisy labels. (iv) 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/breastcancerclassifier.