Background: Mammographic density improves the accuracy of breast cancer risk models. However, the use of breast density is limited by subjective assessment, variation across radiologists, and restricted data. A mammography-based deep learning (DL) model may provide more accurate risk prediction. Purpose:To develop a mammography-based DL breast cancer risk model that is more accurate than established clinical breast cancer risk models. Materials and Methods:This retrospective study included 88 994 consecutive screening mammograms in 39 571 women between January 1, 2009, and December 31, 2012. For each patient, all examinations were assigned to either training, validation, or test sets, resulting in 71 689, 8554, and 8751 examinations, respectively. Cancer outcomes were obtained through linkage to a regional tumor registry. By using risk factor information from patient questionnaires and electronic medical records review, three models were developed to assess breast cancer risk within 5 years: a risk-factor-based logistic regression model (RF-LR) that used traditional risk factors, a DL model (image-only DL) that used mammograms alone, and a hybrid DL model that used both traditional risk factors and mammograms. Comparisons were made to an established breast cancer risk model that included breast density version 8 [TC]). Model performance was compared by using areas under the receiver operating characteristic curve (AUCs) with DeLong test (P , .05). Results:The test set included 3937 women, aged 56.20 years 6 10.04. Hybrid DL and image-only DL showed AUCs of 0.70 (95% confidence interval [CI]: 0.66, 0.75) and 0.68 (95% CI: 0.64, 0.73), respectively. RF-LR and TC showed AUCs of 0.67 (95% CI: 0.62, 0.72) and 0.62 (95% CI: 0.57, 0.66), respectively. Hybrid DL showed a significantly higher AUC (0.70) than TC (0.62; P , .001) and RF-LR (0.67; P = .01). Conclusion:Deep learning models that use full-field mammograms yield substantially improved risk discrimination compared with the Tyrer-Cuzick (version 8) model.

The significant advances in magnetic resonance imaging (MRI) hardware and software, sequence design, and postprocessing methods have made diffusion-weighted imaging (DWI) an important part of body MRI protocols and have fueled extensive research on quantitative diffusion outside the brain, particularly in the oncologic setting. In this review, we summarize the most up-to-date information on DWI acquisition and clinical applications outside the brain, as discussed in an ISMRM-sponsored symposium held in April 2015. We first introduce recent advances in acquisition, processing, and quality control; then review scientific evidence in major organ systems; and finally describe future directions.

Improved breast cancer risk models enable targeted screening strategies that achieve earlier detection and less screening harm than existing guidelines. To bring deep learning risk models to clinical practice, we need to further refine their accuracy, validate them across diverse populations, and demonstrate their potential to improve clinical workflows. We developed Mirai, a mammography-based deep learning model designed to predict risk at multiple timepoints, leverage potentially missing risk factor information, and produce predictions that are consistent across mammography machines. Mirai was trained on a large dataset from Massachusetts General Hospital (MGH) in the United States and tested on held-out test sets from MGH, Karolinska University Hospital in Sweden, and Chang Gung Memorial Hospital (CGMH) in Taiwan, obtaining C-indices of 0.76 (95% confidence interval, 0.74 to 0.80), 0.81 (0.79 to 0.82), and 0.79 (0.79 to 0.83), respectively. Mirai obtained significantly higher 5-year ROC AUCs than the Tyrer-Cuzick model (P < 0.001) and prior deep learning models Hybrid DL (P < 0.001) and Image-Only DL (P < 0.001), trained on the same dataset. Mirai more accurately identified high-risk patients than prior methods across all datasets. On the MGH test set, 41.5% (34.4 to 48.5) of patients who would develop cancer within 5 years were identified as high risk, compared with 36.1% (29.1 to 42.9) by Hybrid DL (P = 0.02) and 22.9% (15.9 to 29.6) by the Tyrer-Cuzick model (P < 0.001).

ammographic breast density can mask cancers at mammography and is an independent risk factor for breast cancer (1-3). Legislation mandating patients be notified of mammographic breast density has passed in more than 30 states, and a federal bill is under consideration. Details of state legislation vary, but most states require direct reporting to the patient that breast density can mask cancers at mammography and that the patient may benefit from additional testing. Qualitative assessment of mammographic breast density is subjective and varies widely between radiologists (4-10). In a study of 83 radiologists who assessed breast density, Sprague et al (4) found extreme variation in qualitative density assessment per the Breast Imaging Reporting and Data System (BI-RADS), with 6%-85% of mammograms assessed as either heterogeneously or extremely dense depending on radiologist interpretation. In a study of 34 radiologists, the intraradiologist agreement of density assessments among women who underwent two examinations varied from 62% to 87% (6). Commercially available methods for automated assessment of breast density do exist; however, they yield mixed results in agreement with expert qualitative density assessments, with k scores of 0.32-0.61 (11,12). These methods tend to result in over-or underreporting of breast density when compared with qualitative assessment by radiologists (11,13). A recent study found significant differences in density assessments in the same 4170 women with two software programs (Volpara, Volpara Solutions, Wellington, New Zealand; Quantra, Hologic, Bedford, Mass), with the software programs showing 37% and 51%, respectively, of women had dense breast tissue. In the same set of mammograms, radiologists determined 43% of the women had dense breast tissue (13). Deep learning (DL) has been gaining traction in radiology (12,14-17). Specifically, there has been preliminary work with DL methods to assess breast density (12,18); however, none of these techniques have been implemented in clinical practice, raising questions about clinical acceptance by practicing radiologists and the effect on patient care. In contrast, our purpose was to develop a DL algorithm we could use to reliably assess breast density and to measure the acceptance of its predictions in real-time clinical practice. We hypothesize that DL models can be applied to assess breast density at the same level as experienced breast imagers and that they can be accepted into routine clinical practice.

• BREAST IMAGING M ammography is the only imaging modality shown to reduce breast cancer mortality in randomized trials (1-8). Despite its benefits, challenges include variation in interpretive performance and the scarcity of specialized radiologists (9,10). A recent report of mammography screening performance in U.S. community practice demonstrated that radiologists' diagnostic performance ranged from 66.7% to 98.6% for sensitivity and from 71.2% to 96.9% for specificity (11). False-negative examinations can result in delayed diagnosis, and false-positive examinations can lead to unnecessary procedures, impacting both patient experience and overall costs. Moreover, the ability of specialized radiologists to serve the global population of women eligible for breast cancer screening is limited by workflow inefficiencies (12,13). Both technologic and workflow solutions have been proposed to improve radiologist interpretive performance and efficiency. Computer-aided detection (CAD), which detects and marks suspicious findings on mammograms, aims to improve radiologist sensitivity. Although traditional approaches have not demonstrated improved radiologist performance in sensitivity or specificity in clinical practice (14-16), a more recent deep learning approach to CAD has shown promise in improving sensitivity in a reader study (17). However, this does not address limitations in radiologist specificity or efficiency. Double reading (ie, having two radiologists interpret the same mammogram) has also been implemented to improve radiologist performance. Although some studies demonstrate slight improvements in sensitivity, double reading worsens workflow efficiency and increases false-positive examination results (18,19). We hypothesized that a deep learning model trained to triage mammograms as cancer free can improve radiologist efficiency and specificity without harming sensitivity. Specifically, we trained a model to predict cancer

There is increasing interest in the potential benefits and harms of screening ultrasound to supplement mammographic screening of women with dense breast tissue. We review the current evidence regarding adjunctive screening breast ultrasound (US) and provide a summary for clinicians who counsel patients with dense breasts. We conducted a comprehensive literature review of published clinical trials and observational cohort studies assessing the efficacy of screening handheld US (HHUS) and automated breast US (ABUS) to supplement mammography among women with dense breasts. From a total of 189 peer-reviewed publications on the performance of screening US, 12 studies were relevant to our analysis. The reporting of breast cancer risk factors varied across studies; however, the study populations tended to be at greater than average risk for developing breast cancer. There is consistent evidence that adjunctive screening US detects more invasive cancers compared to mammography alone, but there is currently no evidence of associated long-term breast cancer mortality reduction. The studies also collectively found that US was associated with an additional 11.7–106.6 biopsies/1,000 examinations (Median 52.2), and detected an additional 0.3–7.7 cancers/1,000 examinations (Median 4.2). The associated number of unnecessary breast biopsies resulting from adjunct US screening exceeds that observed with screening mammography alone by approximately 5-fold. Adjunctive screening with ultrasound should also be considered in the context of screening mammography. It is important for clinicians to be aware that improvements in cancer detection in mammographically dense breasts have been achieved with the transition from film to digital mammography, reducing a limitation of film mammography. Clinicians should discuss breast density as one of several important breast cancer risk factors, consider the potential harms of adjunctive screening, and arrive at a shared decision consistent with each woman’s preferences and values.

PURPOSE Accurate risk assessment is essential for the success of population screening programs in breast cancer. Models with high sensitivity and specificity would enable programs to target more elaborate screening efforts to high-risk populations, while minimizing overtreatment for the rest. Artificial intelligence (AI)-based risk models have demonstrated a significant advance over risk models used today in clinical practice. However, the responsible deployment of novel AI requires careful validation across diverse populations. To this end, we validate our AI-based model, Mirai, across globally diverse screening populations. METHODS We collected screening mammograms and pathology-confirmed breast cancer outcomes from Massachusetts General Hospital, USA; Novant, USA; Emory, USA; Maccabi-Assuta, Israel; Karolinska, Sweden; Chang Gung Memorial Hospital, Taiwan; and Barretos, Brazil. We evaluated Uno's concordance-index for Mirai in predicting risk of breast cancer at one to five years from the mammogram. RESULTS A total of 128,793 mammograms from 62,185 patients were collected across the seven sites, of which 3,815 were followed by a cancer diagnosis within 5 years. Mirai obtained concordance indices of 0.75 (95% CI, 0.72 to 0.78), 0.75 (95% CI, 0.70 to 0.80), 0.77 (95% CI, 0.75 to 0.79), 0.77 (95% CI, 0.73 to 0.81), 0.81 (95% CI, 0.79 to 0.82), 0.79 (95% CI, 0.76 to 0.83), and 0.84 (95% CI, 0.81 to 0.88) at Massachusetts General Hospital, Novant, Emory, Maccabi-Assuta, Karolinska, Chang Gung Memorial Hospital, and Barretos, respectively. CONCLUSION Mirai, a mammography-based risk model, maintained its accuracy across globally diverse test sets from seven hospitals across five countries. This is the broadest validation to date of an AI-based breast cancer model and suggests that the technology can offer broad and equitable improvements in care.

Our learning curve analysis shows that the model can achieve reasonable results even when trained on a few annotations. We developed a user-friendly interface to the database that allows physicians to easily identify patients with target characteristics and export the matching cohort. This model has the potential to reduce the effort required for analyzing large amounts of data from medical records, and to minimize the cost and time required to glean scientific insight from these data.