Search

PurposeThe aim of this study was to develop a prioritization strategy for scheduling immediate screening mammographic interpretation and possible diagnostic evaluation.MethodsA population-based cohort with screening mammograms performed from 2012 to 2020 at 126 radiology facilities from 7Breast Cancer Surveillance Consortium registries was identified. Classification trees identified combinations of clinical history (age, BI-RADS® density, time since prior mammogram, history of false-positive recall or biopsy result), screening modality (digital mammography, digital breast tomosynthesis), and facility characteristics (profit status, location, screening volume, practice type, academic affiliation) that grouped screening mammograms by recall rate, with ≥12/100 considered high and ≥16/100 very high. An efficiency ratio was estimated as the percentage of recalls divided by the percentage of mammograms.ResultsThe study cohort included 2,674,051 screening mammograms in 925,777 women, with 235,569 recalls. The most important predictor of recall was time since prior mammogram, followed by age, history of false-positive recall, breast density, history of benign biopsy, and screening modality. Recall rates were very high for baseline mammograms (21.3/100; 95% confidence interval, 19.7-23.0) and high for women with ≥5 years since prior mammogram (15.1/100; 95% confidence interval, 14.3-16.1). The 9.2% of mammograms in subgroups with very high and high recall rates accounted for 19.2% of recalls, an efficiency ratio of 2.1 compared with a random approach. Adding women

Many factors are involved in the successful development of early career breast imaging radiologists into independent investigators conducting impactful research. Key basic prerequisites for success include a motivated and resilient radiologist, institutional and departmental commitment to supporting early career physician-scientists, strong mentorship, and a flexible strategy for extramural funding that accounts for individualized professional goals. In this review, we describe these factors in greater detail, providing a practical overview for residents, fellows, and junior faculty who are interested in an academic career as a breast imaging radiologist engaged in original scientific research. We also describe the essential pieces of grant applications and summarize the professional milestones for early career physician-scientists as they look toward promotion to associate professor and sustained extramural funding.

BACKGROUND: Artificial intelligence (AI) may improve cancer detection and risk prediction during mammography screening, but radiologists' preferences regarding its characteristics and implementation are unknown. PURPOSE: To quantify how different attributes of AI-based cancer detection and risk prediction tools affect radiologists' intentions to use AI during screening mammography interpretation. MATERIALS AND METHODS: Through qualitative interviews with radiologists, we identified five primary attributes for AI-based breast cancer detection and four for breast cancer risk prediction. We developed a discrete choice experiment based on these attributes and invited 150 US-based radiologists to participate. Each respondent made eight choices for each tool between three alternatives: two hypothetical AI-based tools versus screening without AI. We analyzed samplewide preferences using random parameters logit models and identified subgroups with latent class models. RESULTS: Respondents (n = 66; 44% response rate) were from six diverse practice settings across eight states. Radiologists were more interested in AI for cancer detection when sensitivity and specificity were balanced (94% sensitivity with

Background Breast cancer risk models guide screening and chemoprevention decisions, but the extent and effect of variability among models, particularly at the individual level, is uncertain. Objective To quantify the accuracy and disagreement between commonly used risk models in categorizing individual women as average vs. high risk for developing invasive breast cancer. Design Comparison of three risk prediction models: Breast Cancer Risk Assessment Tool (BCRAT), Breast Cancer Surveillance Consortium (BCSC) model, and International Breast Intervention Study (IBIS) model. Subjects Women 40 to 74 years of age presenting for screening mammography at a multisite health system between 2011 and 2015, with 5-year follow-up for cancer outcome. Main Measures Comparison of model discrimination and calibration at the population level and inter-model agreement for 5-year breast cancer risk at the individual level using two cutoffs (≥ 1.67% and ≥ 3.0%). Key Results A total of 31,115 women were included. When using the ≥ 1.67% threshold, more than 21% of women were classified as high risk for developing breast cancer in the next 5 years by one model, but average risk by another model. When using the ≥ 3.0% threshold, more than 5% of women had disagreements in risk severity between models. Almost half of the women (46.6%) were classified as high risk by at least one of the three models (e.g., if all three models were applied) for the threshold of ≥ 1.67%, and 11.1% were classified as high risk for ≥ 3.0%. All three models had similar accuracy at the population level. Conclusions Breast cancer risk estimates for individual women vary substantially, depending on which risk assessment model is used. The choice of cutoff used to define high risk can lead to adverse effects for screening, preventive care, and quality of life for misidentified individuals. Clinicians need to be aware of the high false-positive and false-negative rates and variation between models when talking with patients.

ImportanceWith a shortfall in fellowship-trained breast radiologists, mammography screening programs are looking toward artificial intelligence (AI) to increase efficiency and diagnostic accuracy. External validation studies provide an initial assessment of how promising AI algorithms perform in different practice settings.ObjectiveTo externally validate an ensemble deep-learning model using data from a high-volume, distributed screening program of an academic health system with a diverse patient population.Design, Setting, and ParticipantsIn this diagnostic study, an ensemble learning method, which reweights outputs of the 11 highest-performing individual AI models from the Digital Mammography Dialogue on Reverse Engineering Assessment and Methods (DREAM) Mammography Challenge, was used to predict the cancer status of an individual using a standard set of screening mammography images. This study was conducted using retrospective patient data collected between 2010 and 2020 from women aged 40 years and older who underwent a routine breast screening examination and participated in the Athena Breast Health Network at the University of California, Los Angeles (UCLA).Main Outcomes and MeasuresPerformance of the challenge ensemble method (CEM) and the CEM combined with radiologist assessment (CEM+R) were compared with diagnosed ductal carcinoma in situ and invasive cancers within a year of the screening examination using performance metrics, such as sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC).ResultsEvaluated on 37 317 examinations from 26 817 women (mean [SD] age, 58.4 [11.5] years), individual model AUROC estimates ranged from 0.77 (95% CI, 0.75-0.79) to 0.83 (95% CI, 0.81-0.85). The CEM model achieved an AUROC of 0.85 (95% CI, 0.84-0.87) in the UCLA cohort, lower than the performance achieved in the Kaiser Permanente Washington (AUROC, 0.90) and Karolinska Institute (AUROC, 0.92) cohorts. The CEM+R model achieved a sensitivity (0.813 [95% CI, 0.781-0.843] vs 0.826 [95% CI, 0.795-0.856]; P = .20) and specificity (0.925 [95% CI, 0.916-0.934] vs 0.930 [95% CI, 0.929-0.932]; P = .18) similar to the radiologist performance. The CEM+R model had significantly lower sensitivity (0.596 [95% CI, 0.466-0.717] vs 0.850 [95% CI, 0.766-0.923]; P P P = .004).Conclusions and RelevanceThis study found that the high performance of an ensemble deep-learning model for automated screening mammography interpretation did not generalize to a more diverse screening cohort, suggesting that the model experienced underspecification. This study suggests the need for model transparency and fine-tuning of AI models for specific target populations prior to their clinical adoption.

Background: The purpose of this study was to determine factors associated with receipt of screening mammography by insured women before breast cancer diagnosis, and subsequent outcomes. Patients and Methods: Using claims data from commercial and federal payers linked to a regional SEER registry, we identified women diagnosed with breast cancer from 2007 to 2017 and determined receipt of screening mammography within 1 year before diagnosis. We obtained patient and tumor characteristics from the SEER registry and assigned each woman a socioeconomic deprivation score based on residential address. Multivariable logistic regression models were used to evaluate associations of patient and tumor characteristics with late-stage disease and nonreceipt of mammography. We used multivariable Cox proportional hazards models to identify predictors of subsequent mortality. Results: Among 7,047 women, 69% (n=4,853) received screening mammography before breast cancer diagnosis. Compared with women who received mammography, those with no mammography had a higher proportion of late-stage disease (34% vs 10%) and higher 5-year mortality (18% vs 6%). In multivariable modeling, late-stage disease was most associated with nonreceipt of mammography (odds ratio [OR], 4.35; 95% CI, 3.80–4.98). The Cox model indicated that nonreceipt of mammography predicted increased risk of mortality (hazard ratio [HR], 2.00; 95% CI, 1.64–2.43), independent of late-stage disease at diagnosis (HR, 5.00; 95% CI, 4.10–6.10), Charlson comorbidity index score ≥1 (HR, 2.75; 95% CI, 2.26–3.34), and negative estrogen receptor/progesterone receptor status (HR, 2.09; 95% CI, 1.67–2.61). Nonreceipt of mammography was associated with younger age (40–49 vs 50–59 years; OR, 1.69; 95% CI, 1.45–1.96) and increased socioeconomic deprivation (OR, 1.05 per decile increase; 95% CI, 1.03–1.07). Conclusions: In a cohort of insured women diagnosed with breast cancer, nonreceipt of screening mammography was significantly associated with late-stage disease and mortality, suggesting that interventions to further increase uptake of screening mammography may improve breast cancer outcomes.

Background Artificial intelligence (AI) is increasingly being proposed for use in medicine, including breast cancer screening (BCS). Little is known, however, about referring primary care providers’ (PCPs’) preferences for this technology. Methods We identified the most important attributes of AI BCS for ordering PCPs using qualitative interviews: sensitivity, specificity, radiologist involvement, understandability of AI decision-making, supporting evidence, and diversity of training data. We invited US-based PCPs to participate in an internet-based experiment designed to force participants to trade off among the attributes of hypothetical AI BCS products. Responses were analyzed with random parameters logit and latent class models to assess how different attributes affect the choice to recommend AI-enhanced screening. Results Ninety-one PCPs participated. Sensitivity was most important, and most PCPs viewed radiologist participation in mammography interpretation as important. Other important attributes were specificity, understandability of AI decision-making, and diversity of data. We identified 3 classes of respondents: “Sensitivity First” (41%) found sensitivity to be more than twice as important as other attributes; “Against AI Autonomy” (24%) wanted radiologists to confirm every image; “Uncertain Trade-Offs” (35%) viewed most attributes as having similar importance. A majority (76%) accepted the use of AI in a “triage” role that would allow it to filter out likely negatives without radiologist confirmation. Conclusions and Relevance Sensitivity was the most important attribute overall, but other key attributes should be addressed to produce clinically acceptable products. We also found that most PCPs accept the use of AI to make determinations about likely negative mammograms without radiologist confirmation.

Background: Social determinants of health (SDOH) contribute to health care disparities, with social and economic barriers often leading to difficulties in obtaining necessary care. We evaluated barriers to receiving health care, focusing on caretaker responsibilities, health insurance and cost, and transportation. Materials and Methods: We included women ages ≥40 years receiving screening mammography across three Breast Cancer Surveillance Consortium registries from 2012 to 2017. Women self-reported social and financial barriers to receiving health care in the 12 months before their screening mammogram. We evaluated woman- and census-based community-level factors associated with reporting a barrier using multivariate logistic regression. We assessed interaction with urban versus nonurban residence using Wald tests. Results: Among 393,430 women, 3.6% reported a barrier with a higher proportion in urban versus nonurban settings (3.9% [n = 11,977] vs. 2.2% [n = 1,655], respectively; p

Background Established mammography screening performance metrics use the initial screening mammography assessment because they were developed for radiologist performance auditing, yet these metrics are frequently used to inform health policy and screening decision making. The authors have developed new performance metrics based on the final assessment that consider the entire screening episode, including diagnostic workup. Methods The authors used data from 2,512,577 screening episodes during 2005-2017 at 146 facilities in the United States participating in the Breast Cancer Surveillance Consortium. Screening performance metrics based on the final assessment of the screening episode were compared with conventional metrics defined with the initial assessment. Results were also stratified by breast density and breast cancer risk. Results The cancer detection rates were similar for the final assessment (4.1 per 1000; 95% confidence interval [CI], 3.8-4.3 per 1000) and the initial assessment (4.1 per 1000; 95% CI, 3.9-4.3 per 1000). The interval cancer rate was 12% higher when it was based on the final assessment (0.77 per 1000; 95% CI, 0.71-0.83 per 1000) versus the initial assessment (0.69 per 1000; 95% CI, 0.64-0.74 per 1000), and this resulted in a modest difference in sensitivity (84.1% [95% CI, 83.0%-85.1%] vs 85.7% [95% CI, 84.8%-86.6%], respectively). Absolute differences in the interval cancer rate between final and initial assessments increased with breast density and breast cancer risk (eg, a difference of 0.29 per 1000 for women with extremely dense breasts and a 5-year risk >2.49%). Conclusions Established screening performance metrics underestimate the interval cancer rate of a mammography screening episode, particularly for women with dense breasts or an elevated breast cancer risk. Women, clinicians, policymakers, and researchers should use final-assessment performance metrics to support informed screening decisions.

Objectives To analyze rates, odds ratios (OR), and characteristics of screen-detected and interval cancers after concordant and discordant initial interpretations and consensus in a population-based screening program. Methods Data were extracted from the Cancer Registry of Norway for 487,118 women who participated in BreastScreen Norway, 2006–2017, with 2 years of follow-up. All mammograms were independently interpreted by two radiologists, using a score from 1 (negative) to 5 (high suspicion of cancer). A score of 2+ by one of the two radiologists was defined as discordant and 2+ by both radiologists as concordant positive. Consensus was performed on all discordant and concordant positive, with decisions of recall for further assessment or dismiss. OR was estimated with logistic regression with 95% confidence interval (CI), and histopathological tumor characteristics were analyzed for screen-detected and interval cancer. Results Among screen-detected cancers, 23.0% (697/3024) had discordant scores, while 12.8% (117/911) of the interval cancers were dismissed at index screening. Adjusted OR was 2.4 (95% CI: 1.9–2.9) for interval cancer and 2.8 (95% CI: 2.5–3.2) for subsequent screen-detected cancer for women dismissed at consensus compared to women with concordant negative scores. We found 3.4% (4/117) of the interval cancers diagnosed after being dismissed to be DCIS, compared to 20.3% (12/59) of those with false-positive result after index screening. Conclusion Twenty-three percent of the screen-detected cancers was scored negative by one of the two radiologists. A higher odds of interval and subsequent screen-detected cancer was observed among women dismissed at consensus compared to concordant negative scores. Our findings indicate a benefit of personalized follow-up. Key Points • In this study of 487,118 women participating in a screening program using independent double reading with consensus, 23% screen-detected cancers were detected by only one of the two radiologists. • The adjusted odds ratio for interval cancer was 2.4 (95% confidence interval: 1.9, 2.9) for cases dismissed at consensus using concordant negative interpretations as the reference. • Interval cancers diagnosed after being dismissed at consensus or after concordant negative scores had clinically less favorable prognostic tumor characteristics compared to those diagnosed after false-positive results.

Breast cancer screening with digital breast tomosynthesis may decrease false-positive results compared with digital mammography.To estimate the probability of receiving at least 1 false-positive result after 10 years of screening with digital breast tomosynthesis vs digital mammography in the US.An observational comparative effectiveness study with data collected prospectively for screening examinations was performed between January 1, 2005, and December 31, 2018, at 126 radiology facilities in the Breast Cancer Surveillance Consortium. Analysis included 903 495 individuals aged 40 to 79 years. Data analysis was conducted from February 9 to September 7, 2021.Screening modality, screening interval, age, and Breast Imaging Reporting and Data System breast density.Cumulative risk of at least 1 false-positive recall for further imaging, short-interval follow-up recommendation, and biopsy recommendation after 10 years of annual or biennial screening with digital breast tomosynthesis vs digital mammography, accounting for competing risks of breast cancer diagnosis and death.In this study of 903 495 women, 2 969 055 nonbaseline screening examinations were performed with interpretation by 699 radiologists. Mean (SD) age of the women at the time of the screening examinations was 57.6 (9.9) years, and 58% of the examinations were in individuals younger than 60 years and 46% were performed in women with dense breasts. A total of 15% of examinations used tomosynthesis. For annual screening, the 10-year cumulative probability of at least 1 false-positive result was significantly lower with tomosynthesis vs digital mammography for all outcomes: 49.6% vs 56.3% (difference, -6.7; 95% CI, -7.4 to -6.1) for recall, 16.6% vs 17.8% (difference, -1.1; 95% CI, -1.7 to -0.6) for short-interval follow-up recommendation, and 11.2% vs 11.7% (difference, -0.5; 95% CI, -1.0 to -0.1) for biopsy recommendation. For biennial screening, the cumulative probability of a false-positive recall was significantly lower for tomosynthesis vs digital mammography (35.7% vs 38.1%; difference, -2.4; 95% CI, -3.4 to -1.5), but cumulative probabilities did not differ significantly by modality for short-interval follow-up recommendation (10.3% vs 10.5%; difference, -0.1; 95% CI, -0.7 to 0.5) or biopsy recommendation (6.6% vs 6.7%; difference, -0.1; 95% CI, -0.5 to 0.4). Decreases in cumulative probabilities of false-positive results with tomosynthesis vs digital mammography were largest for annual screening in women with nondense breasts (differences for recall, -6.5 to -12.8; short-interval follow-up, 0.1 to -5.2; and biopsy recommendation, -0.5 to -3.1). Regardless of modality, cumulative probabilities of false-positive results were substantially lower for biennial vs annual screening (overall recall, 35.7 to 38.1 vs 49.6 to 56.3; short-interval follow-up, 10.3 to 10.5 vs 16.6 to 17.8; and biopsy recommendation, 6.6 to 6.7 vs 11.2 to 11.7); older vs younger age groups (eg, among annual screening in women ages 70-79 vs 40-49, recall, 39.8 to 47.0 vs 60.8 to 68.0; short-interval follow-up, 13.3 to 14.2 vs 20.7 to 20.9; and biopsy recommendation, 9.1 to 9.3 vs 13.2 to 13.4); and women with entirely fatty vs extremely dense breasts (eg, among annual screening in women aged 50-59 years, recall, 29.1 to 36.3 vs 58.8 to 60.4; short-interval follow-up, 8.9 to 11.6 vs 19.5 to 19.8; and biopsy recommendation, 4.9 to 8.0 vs 15.1 to 15.3).In this comparative effectiveness study, 10-year cumulative probabilities of false-positive results were lower on digital breast tomosynthesis vs digital mammography. Biennial screening interval, older age, and nondense breasts were associated with larger reductions in false-positive probabilities than screening modality.

Purpose: The aim of this study was to describe the current state of science regarding independent external validation of artificial intelligence (AI) technologies for screening mammography. Methods: A systematic review was performed across five databases (Embase, PubMed, IEEE Explore, Engineer Village, and arXiv) through December 10, 2020. Studies that used screening examinations from real-world settings to externally validate AI algorithms for mammographic cancer detection were included. The main outcome was diagnostic accuracy, defined by area under the receiver operating characteristic curve (AUC). Performance was also compared between radiologists and either stand-alone AI or combined radiologist and AI interpretation. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Results: After data extraction, 13 studies met the inclusion criteria (148,361 total patients). Most studies (77% [n = 10]) evaluated commercially available AI algorithms. Studies included retrospective reader studies (46% [n = 6]), retrospective simulation studies (38% [n = 5]), or both (15% [n = 2]). Across 5 studies comparing stand-alone AI with radiologists, 60% (n = 3) demonstrated improved accuracy with AI (AUC improvement range, 0.02-0.13). All 5 studies comparing combined radiologist and AI interpretation with radiologists alone demonstrated improved accuracy with AI (AUC improvement range, 0.028-0.115). Most studies had risk for bias or applicability concerns for patient selection (69% [n = 9]) and the reference standard (69% [n = 9]). Only two studies obtained ground-truth cancer outcomes through regional cancer registry linkage. Conclusions: To date, external validation efforts for AI screening mammographic technologies suggest small potential diagnostic accuracy improvements but have been retrospective in nature and suffer from risk for bias and applicability concerns.

The performance of the artificial intelligence system was promising for breast cancer detection in a large population-based mammography screening program. Background Artificial intelligence (AI) has shown promising results for cancer detection with mammographic screening. However, evidence related to the use of AI in real screening settings remain sparse. Purpose To compare the performance of a commercially available AI system with routine, independent double reading with consensus as performed in a population-based screening program. Furthermore, the histopathologic characteristics of tumors with different AI scores were explored. Materials and Methods In this retrospective study, 122 969 screening examinations from 47 877 women performed at four screening units in BreastScreen Norway from October 2009 to December 2018 were included. The data set included 752 screen-detected cancers (6.1 per 1000 examinations) and 205 interval cancers (1.7 per 1000 examinations). Each examination had an AI score between 1 and 10, where 1 indicated low risk of breast cancer and 10 indicated high risk. Threshold 1, threshold 2, and threshold 3 were used to assess the performance of the AI system as a binary decision tool (selected vs not selected). Threshold 1 was set at an AI score of 10, threshold 2 was set to yield a selection rate similar to the consensus rate (8.8%), and threshold 3 was set to yield a selection rate similar to an average individual radiologist (5.8%). Descriptive statistics were used to summarize screening outcomes. Results A total of 653 of 752 screen-detected cancers (86.8%) and 92 of 205 interval cancers (44.9%) were given a score of 10 by the AI system (threshold 1). Using threshold 3, 80.1% of the screen-detected cancers (602 of 752) and 30.7% of the interval cancers (63 of 205) were selected. Screen-detected cancer with AI scores not selected using the thresholds had favorable histopathologic characteristics compared to those selected; opposite results were observed for interval cancer. Conclusion The proportion of screen-detected cancers not selected by the artificial intelligence (AI) system at the three evaluated thresholds was less than 20%. The overall performance of the AI system was promising according to cancer detection.

IntroductionArtificial intelligence (AI) algorithms for interpreting mammograms have the potential to improve the effectiveness of population breast cancer screening programmes if they can detect cancers, including interval cancers, without contributing substantially to overdiagnosis. Studies suggesting that AI has comparable or greater accuracy than radiologists commonly employ ‘enriched’ datasets in which cancer prevalence is higher than in population screening. Routine screening outcome metrics (cancer detection and recall rates) cannot be estimated from these datasets, and accuracy estimates may be subject to spectrum bias which limits generalisabilty to real-world screening. We aim to address these limitations by comparing the accuracy of AI and radiologists in a cohort of consecutive of women attending a real-world population breast cancer screening programme.Methods and analysisA retrospective, consecutive cohort of digital mammography screens from 109 000 distinct women was assembled from BreastScreen WA (BSWA), Western Australia’s biennial population screening programme, from November 2016 to December 2017. The cohort includes 761 screen-detected and 235 interval cancers. Descriptive characteristics and results of radiologist double-reading will be extracted from BSWA outcomes data collection. Mammograms will be reinterpreted by a commercial AI algorithm (DeepHealth). AI accuracy will be compared with that of radiologist single-reading based on the difference in the area under the receiver operating characteristic curve. Cancer detection and recall rates for combined AI–radiologist reading will be estimated by pairing the first radiologist read per screen with the AI algorithm, and compared with estimates for radiologist double-reading.Ethics and disseminationThis study has ethical approval from the Women and Newborn Health Service Ethics Committee (EC00350) and the Curtin University Human Research Ethics Committee (HRE2020-0316). Findings will be published in peer-reviewed journals and presented at national and international conferences. Results will also be disseminated to stakeholders in Australian breast cancer screening programmes and policy makers in population screening.

Background: We evaluated differences in diagnostic mammography performance based on women's race/ethnicity. Methods: This cohort study included 267,868 diagnostic mammograms performed to evaluate screening mammogram findings at 98 facilities in the Breast Cancer Surveillance Consortium between 2005 and 2017. Mammogram assessments were recorded prospectively and breast cancers occurring within one year were ascertained. Performance statistics were calculated with 95% confidence intervals (CI) for each racial/ethnic group. Multivariable regression was used to control for personal characteristics and imaging facility. Results: Among non-Hispanic White (70%), non-Hispanic Black (13%), Asian/Pacific Islander (10%), and Hispanic (7%) women, the invasive cancer detection rate (iCDR, per 1,000 mammograms) and positive predictive value (PPV2) were highest among non-Hispanic White women (iCDR, 35.8; 95% CI, 35.0–36.7; PPV2, 27.8; 95% CI, 27.3–28.3) and lowest among Hispanic women (iCDR, 22.3; 95% CI, 20.2–24.6; PPV2, 19.4; 95% CI, 18.0–20.9). Short interval follow-up recommendations were most common among non-Hispanic Black women [(31.0%; 95% CI, 30.6%–31.5%) vs. other groups, range, 16.6%–23.6%]. False-positive biopsy recommendations were most common among Asian/Pacific Islander women [per 1,000 mammograms: 169.2; 95% CI, 164.8–173.7) vs. other groups, range, 126.5–136.1]. Some differences were explained by adjusting for receipt of diagnostic ultrasound or MRI for iCDR and imaging facility for short-interval follow-up. Other differences changed little after adjustment. Conclusions: Diagnostic mammography performance varied across racial/ethnic groups. Addressing characteristics related to imaging facility and access, rather than personal characteristics, may help reduce some of these disparities. Impact: Diagnostic mammography performance studies should include racially and ethnically diverse populations to provide an accurate view of the population-level effects.

To compare academic and demographic metrics among recipients of three major early career radiology, interventional radiology, and radiation oncology grants to National Institutes of Health (NIH) K awardees at the time the grants were awarded and then over the course of their careers.Radiologists who received the RSNA Research Scholar Grant, General Electric Radiology Research Academic Fellowship (GERRAF), American Roentgen Ray Society (ARRS) Scholar Award, or NIH Career Development (K) Award before January 1, 2015, were included. Research metrics at the time of grant award (eg, publications) and subsequent scholarly productivity (eg, academic rank, h-index, NIH funding) were recorded until April 2020. Wilcoxon ranked-sum, χThere were 279 recipients: 48 K Award, 115 RSNA Research Scholar Grant, 36 ARRS, and 80 GERRAF. At the time of grant awarding, GERRAF recipients were less likely to have an MD-PhD degree (odds ratio [OR]: 0.36; P = .002) and were more likely to be women (OR: 1.55; P = .042) than K Award recipients. Similarly, recipients of the ARRS (OR: 2.87; P = .010) and GERRAF (OR: 3.19; P = .002) were more likely to have a master's degree. Academic rank, leadership positions, and R01 funding were significant predictors of h-index and total publications over time. Academic rank and the GERRAF were significant predictors of subsequent NIH funding duration but there were no significant predictors of NIH funding amount.Early career radiology awards, specifically the GERRAF, provide support for female and non-PhD investigators and result in comparable academic performance metrics to NIH K Award recipients.

This chapter provides a summary of a landmark study in radiology. How much radiation do children receive from computed tomography and what is the cancer risk from this radiation exposure? Starting with that question, it describes the basics of the study, including funding, study location, who was studied, how many patients, study overview, study intervention, follow-up, endpoints, results, and criticism and limitations. The chapter briefly reviews other relevant studies and information, discusses implications, and concludes with a relevant clinical case.

This chapter provides a summary of a landmark study in radiology. What is the impact of early imaging for back pain among older adults? Starting with that question, it describes the basics of the study, including funding, study location, who was studied, how many patients, study overview, study intervention, follow-up, endpoints, results, and criticism and limitations. The chapter briefly reviews other relevant studies and information, discusses implications, and concludes with a relevant clinical case.

Key Points Question What can be learned from a crowdsourced challenge for the prediction of COVID-19 diagnosis and hospitalization? Findings This diagnostic and prognostic study used a model-to-data approach to implement a continuous benchmarking challenge that has enabled 482 participants to join in the effort to use regularly updated COVID-19 patient data to build machine learning models for COVID-19 diagnosis and hospitalization prediction. Machine learning models showed high accuracy in COVID-19 outcome prediction, but analysis of subgroups and prospective data revealed limitations and bias in the models. Meaning This study suggests that crowdsourced clinical algorithms can predict COVID-19 diagnosis and hospitalization, but evaluation of the submitted models using reserved data sets is necessary to avoid self-assessment traps., Importance Machine learning could be used to predict the likelihood of diagnosis and severity of illness. Lack of COVID-19 patient data has hindered the data science community in developing models to aid in the response to the pandemic. Objectives To describe the rapid development and evaluation of clinical algorithms to predict COVID-19 diagnosis and hospitalization using patient data by citizen scientists, provide an unbiased assessment of model performance, and benchmark model performance on subgroups. Design, Setting, and Participants This diagnostic and prognostic study operated a continuous, crowdsourced challenge using a model-to-data approach to securely enable the use of regularly updated COVID-19 patient data from the University of Washington by participants from May 6 to December 23, 2020. A postchallenge analysis was conducted from December 24, 2020, to April 7, 2021, to assess the generalizability of models on the cumulative data set as well as subgroups stratified by age, sex, race, and time of COVID-19 test. By December 23, 2020, this challenge engaged 482 participants from 90 teams and 7 countries. Main Outcomes and Measures Machine learning algorithms used patient data and output a score that represented the probability of patients receiving a positive COVID-19 test result or being hospitalized within 21 days after receiving a positive COVID-19 test result. Algorithms were evaluated using area under the receiver operating characteristic curve (AUROC) and area under the precision recall curve (AUPRC) scores. Ensemble models aggregating models from the top challenge teams were developed and evaluated. Results In the analysis using the cumulative data set, the best performance for COVID-19 diagnosis prediction was an AUROC of 0.776 (95% CI, 0.775-0.777) and an AUPRC of 0.297, and for hospitalization prediction, an AUROC of 0.796 (95% CI, 0.794-0.798) and an AUPRC of 0.188. Analysis on top models submitting to the challenge showed consistently better model performance on the female group than the male group. Among all age groups, the best performance was obtained for the 25- to 49-year age group, and the worst performance was obtained for the group aged 17 years or younger. Conclusions and Relevance In this diagnostic and prognostic study, models submitted by citizen scientists achieved high performance for the prediction of COVID-19 testing and hospitalization outcomes. Evaluation of challenge models on demographic subgroups and prospective data revealed performance discrepancies, providing insights into the potential bias and limitations in the models., This diagnostic and prognostic study describes the development and evaluation of clinical algorithms to predict COVID-19 diagnosis and hospitalization using patient data by citizen scientists, provides an assessment of model performance, and benchmarks model performance on subgroups.

The practice leaders shared common concerns with different approaches in the immediate and longer terms. By the time this information is available to the reader, the projections for the disease burden and the recovery will have changed again. Already, some states and institutions are reporting a plateau in COVID-19 cases, and non-COVID-related admissions are on the rise. Nevertheless, the practice-level maneuvers to manage extraordinary conditions remain salient as we, collectively, will be systemically stressed again. Heeding these lessons may lessen the impact of second waves of regional COVID-19 infections and future pandemics.

As more artificial intelligence (AI)-enhanced mammography screening tools enter the clinical market, greater focus will be placed on external validation in diverse patient populations. In this viewpoint, we outline lessons learned from prior efforts in this field, the need to validate algorithms on newer screening technologies and diverse patient populations, and conclude by discussing the need for a framework for continuous monitoring and recalibration of these AI tools. Sufficient validation and continuous monitoring of emerging AI tools for breast cancer screening will require greater stakeholder engagement and the creation of shared policies and guidelines.

Background Several European screening mammography programs that use independent double reading with consensus recommend an annual minimum reading volume of 5000 mammograms per radiologist. This recommendation is based only on expert opinion, and the influence of reading volume on performance in such programs is unknown. Purpose To examine the influence of annual and cumulative reading volume on radiologists' reading performance for digital mammography in a screening program that uses independent double reading with consensus. Materials and Methods This retrospective study included data from digital mammographic examinations in BreastScreen Norway obtained from 2006 to 2016. Multilevel mixed-effects models were used to determine how sensitivity, rate of screening-detected breast cancer (SDC), and false-positive rate (FPR) before and after consensus meeting related to annual and cumulative reading volume. Results The study included 2 373 433 readings performed by 121 radiologists. The median annual reading volume ranged from 153 to 19 500 mammograms, and the median cumulative reading volume was 30 566 mammograms. Sensitivity and SDC rate were relatively stable at 87% to 90% and 4.9 to 4.7 per 1000 readings (0.49% and 0.47%), respectively, between 100 and 10 000 annual readings and at 88% to 89% and 4.8 to 4.9 per 1000 readings (0.48% and 0.49%) between 500 and 100 000 cumulative readings. There was a decreasing trend with higher annual volumes (P for trend < .001 for both) to a sensitivity of 81% and SDC rate of 3.9 per 1000 readings (0.39%) at 18 000 readings. There was a decreasing trend in FPR before and after consensus with higher annual and cumulative volumes (P for trend < .001 for all). FPRs before consensus meeting were 5.3% at 100 annual readings and 4.0% at 4000 annual readings and were 6.7% at 500 cumulative readings and 3.6% at 20 000 cumulative readings. FPRs after consensus meeting were 2.5% at 100 annual readings and 2.3% at 4000 annual readings and were 2.7% at 500 cumulative readings and 2.2% at 20 000 cumulative readings. Conclusion Annual reading volumes between 4000 and 10 000 mammograms and cumulative reading volumes greater than 20 000 mammograms may be the most optimal volumes for achieving high reading performance in a screening program with independent double reading. © RSNA, 2019 See also the editorial by Rosenberg and Seidenwurm in this issue.

Objective We conducted a national survey to understand the impact of state-level density reporting laws on women’s level of density risk awareness and their engagement in conversations with providers regarding supplemental screening. Methods In all, 1,500 US women aged 40 to 74 years who obtained a mammogram within 2 years were surveyed in February 2018. The sampling design yielded 300 respondents in each of five groups categorized based on density reporting law features. Women were asked about their breast density-related knowledge, importance of being notified, and sources of information and if conversations with providers regarding density and supplemental screening occurred. Survey results were compared across groups and between women residing in states with versus without density laws. Results The majority of respondents in all groups felt that it is important for women to know their breast density type (range, 85%-90%). Women were most likely informed of breast density type by a health care provider (range, 68%-72%), followed by the mammography result letter (range, 48%-68%), and then a radiologist (range, 46%-61%). Women from states with a density law were significantly more likely to have learned of their breast tissue type from a mammogram results letter (60% versus 48%, P = .011) and discuss supplemental screening (67% versus 53%, respectively; P = .008) than women from states without a law. Conclusion State-level density reporting laws are associated with increased breast density awareness and increased likelihood of conversations between women and their providers regarding supplemental screening.

ImportanceDiagnostic delays in breast cancer detection may be associated with later-stage disease and higher anxiety, but data on multilevel factors associated with diagnostic delay are limited.ObjectiveTo evaluate individual-, neighborhood-, and health care–level factors associated with differences in time from abnormal screening to biopsy among racial and ethnic groups.Design, Setting, and ParticipantsThis prospective cohort study used data from women aged 40 to 79 years who had abnormal results in screening mammograms conducted in 109 imaging facilities across 6 US states between 2009 and 2019. Data were analyzed from February 21 to November 4, 2021.ExposuresIndividual-level factors included self-reported race and ethnicity, age, family history of breast cancer, breast density, previous breast biopsy, and time since last mammogram; neighborhood-level factors included geocoded education and income based on residential zip codes and rurality; and health care–level factors included mammogram modality, screening facility academic affiliation, and facility onsite biopsy service availability. Data were also assessed by examination year.Main Outcome and MeasuresThe main outcome was unadjusted and adjusted relative risk (RR) of no biopsy within 30, 60, and 90 days using sequential log-binomial regression models. A secondary outcome was unadjusted and adjusted median time to biopsy using accelerated failure time models.ResultsA total of 45 186 women (median [IQR] age at screening, 56 [48-65] years) with 46 185 screening mammograms with abnormal results were included. Of screening mammograms with abnormal results recommended for biopsy, 15 969 (34.6%) were not resolved within 30 days, 7493 (16.2%) were not resolved within 60 days, and 5634 (12.2%) were not resolved within 90 days. Compared with White women, there was increased risk of no biopsy within 30 and 60 days for Asian (30 days: RR, 1.66; 95% CI, 1.31-2.10; 60 days: RR, 1.58; 95% CI, 1.15-2.18), Black (30 days: RR, 1.52; 95% CI, 1.30-1.78; 60 days: 1.39; 95% CI, 1.22-1.60), and Hispanic (30 days: RR, 1.50; 95% CI, 1.24-1.81; 60 days: 1.38; 95% CI, 1.11-1.71) women; however, the unadjusted risk of no biopsy within 90 days only persisted significantly for Black women (RR, 1.28; 95% CI, 1.11-1.47). Sequential adjustment for selected individual-, neighborhood-, and health care–level factors, exclusive of screening facility, did not substantially change the risk of no biopsy within 90 days for Black women (RR, 1.27; 95% CI, 1.12-1.44). After additionally adjusting for screening facility, the increased risk for Black women persisted but showed a modest decrease (RR, 1.20; 95% CI, 1.08-1.34).Conclusions and RelevanceIn this cohort study involving a diverse cohort of US women recommended for biopsy after abnormal results on screening mammography, Black women were the most likely to experience delays to diagnostic resolution after adjusting for multilevel factors. These results suggest that adjustment for multilevel factors did not entirely account for differences in time to breast biopsy, but unmeasured factors, such as systemic racism and other health care system factors, may impact timely diagnosis.

PURPOSE Preoperative breast MRI is used to evaluate for additional cancer and extent of disease for newly diagnosed breast cancer, yet benefits and harms of preoperative MRI are not well-documented. We examined whether preoperative MRI yields additional biopsy and cancer detection by extent of breast density. METHODS We followed women in the Breast Cancer Surveillance Consortium with an incident breast cancer diagnosed from 2005 to 2017. We quantified breast biopsies and cancers detected within 6 months of diagnosis by preoperative breast MRI receipt, overall and by breast density, accounting for MRI selection bias using inverse probability weighted logistic regression. RESULTS Among 19,324 women with newly diagnosed breast cancer, 28% had preoperative MRI, 11% additional biopsy, and 5% additional cancer detected. Four times as many women with preoperative MRI underwent additional biopsy compared to women without MRI (22.6% v. 5.1%). Additional biopsy rates with preoperative MRI increased with increasing breast density (27.4% for extremely dense compared to 16.2% for almost entirely fatty breasts). Rates of additional cancer detection were almost four times higher for women with v. without MRI (9.9% v. 2.6%). Conditional on additional biopsy, age-adjusted rates of additional cancer detection were lowest among women with extremely dense breasts, regardless of imaging modality (with MRI: 35.0%; 95% CI 27.0-43.0%; without MRI: 45.1%; 95% CI 32.6-57.5%). CONCLUSION For women with dense breasts, preoperative MRI was associated with much higher biopsy rates, without concomitant higher cancer detection. Preoperative MRI may be considered for some women, but selecting women based on breast density is not supported by evidence. TRIAL REGISTRATION ClinicalTrials.gov: NCT02980848; registered 2017.

INTRODUCTION: Limited evidence exists on the impact of age and comorbidity on biopsy rates and findings among older women. MATERIALS AND METHODS: We used data from 170,657 women ages 66–94 enrolled in the United States Breast Cancer Surveillance Consortium (BCSC). We estimated one-year rates of biopsy by type (any, fine-needle aspiration (FNA), core or surgical) and yield of the most invasive biopsy finding (benign, ductal carcinoma in situ (DCIS) and invasive breast cancer) by age and comorbidity. Statistical significance was assessed using Wald statistics comparing coefficients estimated from logistic regression models adjusted for age, comorbidity, BCSC registry, and interaction between age and comorbidity. RESULTS: Of 524,860 screening mammograms, 9830 biopsies were performed following 7930 exams (1.5%) within one year, specifically 5589 core biopsies (1.1%), 3422 (0.7%) surgical biopsies and 819 FNAs (0.2%). Biopsy rates per 1000 screens decreased with age (66–74:15.7, 95%CI:14.8–16.8), 75–84:14.5(13.5–15.6), 85–94:13.2(11.3,15.4), p(trend)=2:19.0 (16.9–21.5), p(trend)