Breast arterial calcifications on mammography: a new marker of cardiovascular risk in asymptomatic middle age women?

Breast Arterial Calcifications on Mammography: A Review of the Literature

Identifying systemic disease with medical imaging studies may improve population health outcomes. Although the pathogenesis of peripheral arterial calcification and coronary artery calcification differ, breast arterial calcification (BAC) on mammography is associated with cardiovascular disease (CVD), a leading cause of death in women. While professional society guidelines on the reporting or management of BAC have not yet been established, and assessment and quantification methods are not yet standardized, the value of reporting BAC is being considered internationally as a possible indicator of subclinical CVD. Furthermore, artificial intelligence (AI) models are being developed to identify and quantify BAC on mammography, as well as to predict the risk of CVD. This review outlines studies evaluating the association of BAC and CVD, introduces the role of preventative cardiology in clinical management, discusses reasons to consider reporting BAC, acknowledges current knowledge gaps and barriers to assessing and reporting calcifications, and provides examples of how AI can be utilized to measure BAC and contribute to cardiovascular risk assessment. Ultimately, reporting BAC on mammography might facilitate earlier mitigation of cardiovascular risk factors in asymptomatic women.

Evaluation of a deep learning-based software to automatically detect and quantify breast arterial calcifications on digital mammogram

PURPOSE

The purpose of this study was to evaluate an artificial intelligence (AI) software that automatically detects and quantifies breast arterial calcifications (BAC).

MATERIALS AND METHODS

Women who underwent both mammography and thoracic computed tomography (CT) from 2009 to 2018 were retrospectively included in this single-center study. Deep learning-based software was used to automatically detect and quantify BAC with a BAC AI score ranging from 0 to 10-points. Results were compared using Spearman correlation test with a previously described BAC manual score based on radiologists' visual quantification of BAC on the mammogram. Coronary artery calcification (CAC) score was manually scored using a 12-point scale on CT. The diagnostic performance of the marked BAC AI score (defined as BAC AI score ≥ 5) for the detection of marked CAC (CAC score ≥ 4) was analyzed in terms of sensitivity, specificity, accuracy and area under the receiver operating characteristic curve (AUC).

RESULTS

A total of 502 women with a median age of 62 years (age range: 42-96 years) were included. The BAC AI score showed a very strong correlation with the BAC manual score (r = 0.83). Marked BAC AI score had 32.7 % sensitivity (37/113; 95 % confidence interval [CI]: 24.2-42.2), 96.1 % specificity (374/389; 95 % CI: 93.7-97.8), 71.2 % positive predictive value (37/52; 95 % CI: 56.9-82.9), 83.1 % negative predictive value (374/450; 95 % CI: 79.3-86.5), and 81.9 % accuracy (411/502; 95 % CI: 78.2-85.1) for the diagnosis of marked CAC. The AUC of the marked BAC AI score for the diagnosis of marked CAC was 0.64 (95 % CI: 0.60-0.69).

CONCLUSION

The automated BAC AI score shows a very strong correlation with manual BAC scoring in this external validation cohort. The automated BAC AI score may be a useful tool to promote the integration of BAC into mammography reports and to improve awareness of a woman's cardiovascular risk status.

Breast Arterial Calcifications on Mammography: Awareness and Reporting Preferences Amongst Referring Physicians in Canada

Purpose: Breast arterial calcifications (BAC) on mammography have been correlated with increased cardiovascular risk. The Canadian Society of Breast Imaging released a position statement on BAC reporting in January 2023. This study evaluates the awareness of the clinical significance of BAC and reporting preferences of referring physicians in Canada. Methods: A 15-question survey was distributed to Canadian physicians who may review mammography results via regional and subspecialty associations and on social media following local institutional ethical approval. Responses were collected over 10 weeks from February to April 2023. Results: Seventy-two complete responses were obtained. We are unable to determine the response rate, given the means of distribution. Only 17% (12/72) of responding physicians were previously aware of the association between BAC and increased cardiovascular risk, and 51% (37/72) preferred the inclusion of BAC in the mammography report. Fifty-six percent (40/72) indicated that BAC reporting would prompt further investigation, and 63% (45/72) would inform patients that their mammogram showed evidence of BAC. Sixty-nine percent (50/72) would find grading of BAC beneficial and 71% (51/72) agreed that there is a need for national guidelines. Conclusion: Less than a quarter of responding Canadian referring physicians were previously aware of the association between BAC and cardiovascular risk, although half of respondents indicated a preference for BAC reporting on mammography. Most participating physicians would inform their patients of the presence of BAC and consider further cardiovascular risk management. There was consensus that a national BAC grading system and clinical management guidelines would be beneficial.

Automated Breast Arterial Calcification Score Is Associated With Cardiovascular Outcomes and Mortality

Background

Breast arterial calcification (BAC) on mammograms has emerged as a biomarker of women's cardiovascular disease (CVD) risk, but there is a lack of quantification tools and clinical outcomes studies.

Objectives

This study assessed the association of BAC (both presence and quantity) with CVD outcomes.

Methods

This single-center, retrospective study included women with a screening mammogram from 2007 to 2016. BAC was quantified using an artificial intelligence-generated score, which was assessed as both a binary and continuous variable. Regression analyses evaluated the association between BAC and mortality and a composite of acute myocardial infarction, heart failure, stroke, and mortality. Analyses were adjusted for age, race, diabetes, smoking, blood pressure, cholesterol, and history of CVD and chronic kidney disease.

Results

A total of 18,092 women were included in this study (mean age 56.8 ± 11.0 years; diabetes [13%], hypertension [36%], hyperlipidemia [40%], and smoking [5%]). BAC was present in 4,223 (23%). Over a median follow-up of 6 years, death occurred in 7.8% and 2.3% of women with and without BAC, respectively. The composite occurred in 12.4% and 4.3% of women with and without BAC, respectively. Compared to those without, women with BAC had adjusted HRs of 1.49 (95% CI: 1.33-1.67) for mortality and 1.56 (95% CI: 1.41-1.72) for the composite. Each 10-point increase in the BAC score was associated with higher risk of mortality (HR: 1.08 [95% CI: 1.06-1.11]) and the composite (HR: 1.08 [95% CI: 1.06-1.10]). BAC was especially predictive of future events among younger women.

Conclusions

BAC is independently associated with mortality and CVD, especially among younger women. Measurement of BAC beyond presence adds incremental risk stratification. Quantifying BAC using an artificial intelligence algorithm is feasible, clinically relevant, and may improve personalized CVD risk stratification.

Deep transfer learning for detection of breast arterial calcifications on mammograms: a comparative study

INTRODUCTION

Breast arterial calcifications (BAC) are common incidental findings on routine mammograms, which have been suggested as a sex-specific biomarker of cardiovascular disease (CVD) risk. Previous work showed the efficacy of a pretrained convolutional network (CNN), VCG16, for automatic BAC detection. In this study, we further tested the method by a comparative analysis with other ten CNNs.

MATERIAL AND METHODS

Four-view standard mammography exams from 1,493 women were included in this retrospective study and labeled as BAC or non-BAC by experts. The comparative study was conducted using eleven pretrained convolutional networks (CNNs) with varying depths from five architectures including Xception, VGG, ResNetV2, MobileNet, and DenseNet, fine-tuned for the binary BAC classification task. Performance evaluation involved area under the receiver operating characteristics curve (AUC-ROC) analysis, F1-score (harmonic mean of precision and recall), and generalized gradient-weighted class activation mapping (Grad-CAM++) for visual explanations.

RESULTS

The dataset exhibited a BAC prevalence of 194/1,493 women (13.0%) and 581/5,972 images (9.7%). Among the retrained models, VGG, MobileNet, and DenseNet demonstrated the most promising results, achieving AUC-ROCs > 0.70 in both training and independent testing subsets. In terms of testing F1-score, VGG16 ranked first, higher than MobileNet (0.51) and VGG19 (0.46). Qualitative analysis showed that the Grad-CAM++ heatmaps generated by VGG16 consistently outperformed those produced by others, offering a finer-grained and discriminative localization of calcified regions within images.

CONCLUSION

Deep transfer learning showed promise in automated BAC detection on mammograms, where relatively shallow networks demonstrated superior performances requiring shorter training times and reduced resources.

RELEVANCE STATEMENT

Deep transfer learning is a promising approach to enhance reporting BAC on mammograms and facilitate developing efficient tools for cardiovascular risk stratification in women, leveraging large-scale mammographic screening programs.

KEY POINTS

• We tested different pretrained convolutional networks (CNNs) for BAC detection on mammograms. • VGG and MobileNet demonstrated promising performances, outperforming their deeper, more complex counterparts. • Visual explanations using Grad-CAM++ highlighted VGG16's superior performance in localizing BAC.

Assessing cardiovascular risk with mammography and non-contrast chest CT: A review of the literature and clinical implications

Coronary artery disease (CAD) is the leading cause of mortality and disability globally. In the United States, about 7.2% of adults aged 20 and older are affected by CAD. However, due to its progression over decades, CAD is often undetected and unnoticed until plaque ruptures. This leads to partial or complete artery blockage, resulting in myocardial infarction. Thus, new screening methods for early detection of CAD are needed to prevent and minimize the morbidity and mortality from CAD. Vascular calcifications seen on mammography and non-contrast chest CT (NCCT) can be used for the early detection of CAD and are an accurate predictor of cardiovascular risk. This paper aims to review the basic epidemiology, pathophysiology, imaging findings, and correlation of long-term cardiovascular outcomes with vascular calcifications on mammography and NCCT.

Detection and quantification of breast arterial calcifications on mammograms: a deep learning approach

OBJECTIVE

Breast arterial calcifications (BAC) are a sex-specific cardiovascular disease biomarker that might improve cardiovascular risk stratification in women. We implemented a deep convolutional neural network for automatic BAC detection and quantification.

METHODS

In this retrospective study, four readers labelled four-view mammograms as BAC positive (BAC+) or BAC negative (BAC-) at image level. Starting from a pretrained VGG16 model, we trained a convolutional neural network to discriminate BAC+ and BAC- mammograms. Accuracy, F1 score, and area under the receiver operating characteristic curve (AUC-ROC) were used to assess the diagnostic performance. Predictions of calcified areas were generated using the generalized gradient-weighted class activation mapping (Grad-CAM++) method, and their correlation with manual measurement of BAC length in a subset of cases was assessed using Spearman ρ.

RESULTS

A total 1493 women (198 BAC+) with a median age of 59 years (interquartile range 52-68) were included and partitioned in a training set of 410 cases (1640 views, 398 BAC+), validation set of 222 cases (888 views, 89 BAC+), and test set of 229 cases (916 views, 94 BAC+). The accuracy, F1 score, and AUC-ROC were 0.94, 0.86, and 0.98 in the training set; 0.96, 0.74, and 0.96 in the validation set; and 0.97, 0.80, and 0.95 in the test set, respectively. In 112 analyzed views, the Grad-CAM++ predictions displayed a strong correlation with BAC measured length (ρ = 0.88, p < 0.001).

CONCLUSION

Our model showed promising performances in BAC detection and in quantification of BAC burden, showing a strong correlation with manual measurements.

CLINICAL RELEVANCE STATEMENT

Integrating our model to clinical practice could improve BAC reporting without increasing clinical workload, facilitating large-scale studies on the impact of BAC as a biomarker of cardiovascular risk, raising awareness on women's cardiovascular health, and leveraging mammographic screening.

KEY POINTS

• We implemented a deep convolutional neural network (CNN) for BAC detection and quantification. • Our CNN had an area under the receiving operator curve of 0.95 for BAC detection in the test set composed of 916 views, 94 of which were BAC+ . • Furthermore, our CNN showed a strong correlation with manual BAC measurements (ρ = 0.88) in a set of 112 views.

Vitamin K antagonists and cardiovascular calcification: A systematic review and meta-analysis

Background

Many patients treated with Vitamin K antagonists (VKA) for anticoagulation have concomitant vascular or valvular calcification. This meta-analysis aimed to evaluate a hypothesis that vascular and valvular calcification is a side-effect of VKA treatment.

Methods

We conducted a systematic literature search to identify studies that reported vascular or valvular calcification in patients treated with VKA. The associations between VKA use and calcification were analyzed with random-effects inverse variance models and reported as odds ratios (OR) and 95% confidence intervals (95% CI). In addition, univariate meta-regression analyses were utilized to identify any effect moderators.

Results

Thirty-five studies were included (45,757 patients; 6,251 VKA users). The median follow-up was 2.3 years [interquartile range (IQR) of 1.2–4.0]; age 66.2 ± 3.6 years (mean ± SD); the majority of participants were males [77% (IQR: 72–95%)]. VKA use was associated with an increased OR for coronary artery calcification [1.21 (1.08, 1.36), p = 0.001], moderated by the duration of treatment [meta-regression coefficient B of 0.08 (0.03, 0.13), p = 0.0005]. Extra-coronary calcification affecting the aorta, carotid artery, breast artery, and arteries of lower extremities, was also increased in VKA treated patients [1.86 (1.43, 2.42), p < 0.00001] and moderated by the author-reported statistical adjustments of the effect estimates [B: −0.63 (−1.19, −0.08), p = 0.016]. The effect of VKA on the aortic valve calcification was significant [3.07 (1.90, 4.96), p < 0.00001]; however, these studies suffered from a high risk of publication bias.

Conclusion

Vascular and valvular calcification are potential side effects of VKA. The clinical significance of these side effects on cardiovascular outcomes deserves further investigation.