Estimation of breast percent density in raw and processed full field digital mammography images via adaptive fuzzy c-means clustering and support vector machine segmentation

Concordant and discordant breast density patterns by different approaches for assessing breast density and breast cancer risk

PURPOSE

To examine the discrepancy in breast density assessments by radiologists, LIBRA software, and AI algorithm and their association with breast cancer risk.

METHODS

Among 74,610 Korean women aged ≥ 34 years, who underwent screening mammography, density estimates obtained from both LIBRA and the AI algorithm were compared to radiologists using BI-RADS density categories (A-D, designating C and D as dense breasts). The breast cancer risks were compared according to concordant or discordant dense breasts identified by radiologists, LIBRA, and AI. Cox-proportional hazards models were used to determine adjusted hazard ratios (aHRs) [95% confidence intervals (CIs)].

RESULTS

During a median follow-up of 9.9 years, 479 breast cancer cases developed. Compared to the reference non-dense breast group, the aHRs (95% CIs) for breast cancer were 2.37 (1.68-3.36) for radiologist-classified dense breasts, 1.30 (1.05-1.62) for LIBRA, and 2.55 (1.84-3.56) for AI. For different combinations of breast density assessment, aHRs (95% CI) for breast cancer were 2.40 (1.69-3.41) for radiologist-dense/LIBRA-non-dense, 11.99 (1.64-87.62) for radiologist-non-dense/LIBRA-dense, and 2.99 (1.99-4.50) for both dense breasts, compared to concordant non-dense breasts. Similar trends were observed with radiologists/AI classification: the aHRs (95% CI) were 1.79 (1.02-3.12) for radiologist-dense/AI-non-dense, 2.43 (1.24-4.78) for radiologist-non-dense/AI-dense, and 3.23 (2.15-4.86) for both dense breasts.

CONCLUSION

The risk of breast cancer was highest in concordant dense breasts. Discordant dense breast cases also had a significantly higher risk of breast cancer, especially when identified as dense by either AI or LIBRA, but not radiologists, compared to concordant non-dense breast cases.

Mammographic Texture versus Conventional Cumulus Measure of Density in Breast Cancer Risk Prediction: A Literature Review

Mammographic textures show promise as breast cancer risk predictors, distinct from mammographic density. Yet, there is a lack of comprehensive evidence to determine the relative strengths as risk predictor of textures and density and the reliability of texture-based measures. We searched the PubMed database for research published up to November 2023, which assessed breast cancer risk associations [odds ratios (OR)] with texture-based measures and percent mammographic density (PMD), and their discrimination [area under the receiver operating characteristics curve (AUC)], using same datasets. Of 11 publications, for textures, six found stronger associations (P < 0.05) with 11% to 508% increases on the log scale by study, and four found weaker associations (P < 0.05) with 14% to 100% decreases, compared with PMD. Risk associations remained significant when fitting textures and PMD together. Eleven of 17 publications found greater AUCs for textures than PMD (P < 0.05); increases were 0.04 to 0.25 by study. Discrimination from PMD and these textures jointly was significantly higher than from PMD alone (P < 0.05). Therefore, different textures could capture distinct breast cancer risk information, partially independent of mammographic density, suggesting their joint role in breast cancer risk prediction. Some textures could outperform mammographic density for predicting breast cancer risk. However, obtaining reliable texture-based measures necessitates addressing various issues. Collaboration of researchers from diverse fields could be beneficial for advancing this complex field.

Preoperative prediction of nodal status using clinical data and artificial intelligence derived mammogram features enabling abstention of sentinel lymph node biopsy in breast cancer

Introduction

Patients with clinically node-negative breast cancer have a negative sentinel lymph node status (pN0) in approximately 75% of cases and the necessity of routine surgical nodal staging by sentinel lymph node biopsy (SLNB) has been questioned. Previous prediction models for pN0 have included postoperative variables, thus defeating their purpose to spare patients non-beneficial axillary surgery. We aimed to develop a preoperative prediction model for pN0 and to evaluate the contribution of mammographic breast density and mammogram features derived by artificial intelligence for de-escalation of SLNB.

Materials and methods

This retrospective cohort study included 755 women with primary breast cancer. Mammograms were analyzed by commercially available artificial intelligence and automated systems. The additional predictive value of features was evaluated using logistic regression models including preoperative clinical variables and radiological tumor size. The final model was internally validated using bootstrap and externally validated in a separate cohort. A nomogram for prediction of pN0 was developed. The correlation between pathological tumor size and the preoperative radiological tumor size was calculated.

Results

Radiological tumor size was the strongest predictor of pN0 and included in a preoperative prediction model displaying an area under the curve of 0.68 (95% confidence interval: 0.63-0.72) in internal validation and 0.64 (95% confidence interval: 0.59-0.69) in external validation. Although the addition of mammographic features did not improve discrimination, the prediction model provided a 21% SLNB reduction rate when a false negative rate of 10% was accepted, reflecting the accepted false negative rate of SLNB.

Conclusion

This study shows that the preoperatively available radiological tumor size might replace pathological tumor size as a key predictor in a preoperative prediction model for pN0. While the overall performance was not improved by mammographic features, one in five patients could be omitted from axillary surgery by applying the preoperative prediction model for nodal status. The nomogram visualizing the model could support preoperative patient-centered decision-making on the management of the axilla.

Effect of testosterone therapy on breast tissue composition and mammographic breast density in trans masculine individuals

BACKGROUND

The effect of gender-affirming testosterone therapy (TT) on breast cancer risk is unclear. This study investigated the association between TT and breast tissue composition and breast tissue density in trans masculine individuals (TMIs).

METHODS

Of the 444 TMIs who underwent chest-contouring surgeries between 2013 and 2019, breast tissue composition was assessed in 425 TMIs by the pathologists (categories of lobular atrophy and stromal composition) and using our automated deep-learning algorithm (% epithelium, % fibrous stroma, and % fat). Forty-two out of 444 TMIs had mammography prior to surgery and their breast tissue density was read by a radiologist. Mammography digital files, available for 25/42 TMIs, were analyzed using the LIBRA software to obtain percent density, absolute dense area, and absolute non-dense area. Linear regression was used to describe the associations between duration of TT use and breast tissue composition or breast tissue density measures, while adjusting for potential confounders. Analyses stratified by body mass index were also conducted.

RESULTS

Longer duration of TT use was associated with increasing degrees of lobular atrophy (p < 0.001) but not fibrous content (p = 0.82). Every 6 months of TT was associated with decreasing amounts of epithelium (exp(β) = 0.97, 95% CI 0.95,0.98, adj p = 0.005) and fibrous stroma (exp(β) = 0.99, 95% CI 0.98,1.00, adj p = 0.05), but not fat (exp(β) = 1.01, 95%CI 0.98,1.05, adj p = 0.39). The effect of TT on breast epithelium was attenuated in overweight/obese TMIs (exp(β) = 0.98, 95% CI 0.95,1.01, adj p = 0.14). When comparing TT users versus non-users, TT users had 28% less epithelium (exp(β) = 0.72, 95% CI 0.58,0.90, adj p = 0.003). There was no association between TT and radiologist's breast density assessment (p = 0.58) or LIBRA measurements (p > 0.05).

CONCLUSIONS

TT decreases breast epithelium, but this effect is attenuated in overweight/obese TMIs. TT has the potential to affect the breast cancer risk of TMIs. Further studies are warranted to elucidate the effect of TT on breast density and breast cancer risk.

Mammographic density changes after neoadjuvant chemotherapy in triple-negative breast cancer: Association with treatment and survival outcome

PURPOSE

To investigate the association of mammographic breast density with treatment and survival outcomes in patients with triple-negative breast cancer (TNBC) undergoing neoadjuvant chemotherapy (NAC).

METHODS

This retrospective study evaluated 306 women with TNBC who underwent NAC followed by surgery between 2010 and 2019. The baseline density and the density changes after NAC were evaluated. Qualitative breast density (a-d) was evaluated using the Breast Imaging Reporting and Data System. Quantitative breast density (%) was evaluated using fully automated software (the Laboratory for Individualized Breast Radiodensity Assessment) in the contralateral breast. Multivariable logistic regression analysis was used to evaluate the association between breast density and pathologic complete response (pCR), stratified by menopausal status. Cox proportional hazard regression analysis was used to evaluate the association among breast density, the development of contralateral breast cancer, and the development of locoregional recurrence and/or distant metastasis.

RESULTS

Contralateral density reduction ≥10 % was independently associated with pCR in premenopausal women (odds ratio [OR], 2.5; p = 0.022) but not in postmenopausal women (OR, 0.9; p = 0.823). During a mean follow-up of 65 months, 10 (3 %) women developed contralateral breast cancer, and 68 (22 %) women developed locoregional recurrences and/or distant metastases. Contralateral density reduction ≥10 % showed no association with the occurrence of contralateral breast cancer (hazard ratio [HR], 3.1; p = 0.308) or with locoregional recurrence and/or distant metastasis (HR, 1.1; p = 0.794).

CONCLUSION

In premenopausal women, a contralateral breast density reduction of ≥10 % after NAC was independently associated with pCR, although it did not translate into improved outcomes.

Breast percent density changes in digital mammography pre- and post-radiotherapy

INTRODUCTION

Breast cancer (BC), the most frequently diagnosed malignancy among women worldwide, presents a public health challenge and affects mortality rates. Breast-conserving therapy (BCT) is a common treatment, but the risk from residual disease necessitates radiotherapy. Digital mammography monitors treatment response by identifying post-operative and radiotherapy tissue alterations, but accurate assessment of mammographic density remains a challenge. This study used OpenBreast to measure percent density (PD), offering insights into changes in mammographic density before and after BCT with radiation therapy.

METHODS

This retrospective analysis included 92 female patients with BC who underwent BCT, chemotherapy, and radiotherapy, excluding those who received hormonal therapy or bilateral BCT. Percent/percentage density measurements were extracted using OpenBreast, an automated software that applies computational techniques to density analyses. Data were analysed at baseline, 3 months, and 15 months post-treatment using standardised mean difference (SMD) with Cohen's d, chi-square, and paired sample t-tests. The predictive power of PD changes for BC was measured based on the receiver operating characteristic (ROC) curve analysis.

RESULTS

The mean age was 53.2 years. There were no significant differences in PD between the periods. Standardised mean difference analysis revealed no significant changes in the SMD for PD before treatment compared with 3- and 15-months post-treatment. Although PD increased numerically after radiotherapy, ROC analysis revealed optimal sensitivity at 15 months post-treatment for detecting changes in breast density.

CONCLUSIONS

This study utilised an automated breast density segmentation tool to assess the changes in mammographic density before and after BC treatment. No significant differences in the density were observed during the short-term follow-up period. However, the results suggest that quantitative density assessment could be valuable for long-term monitoring of treatment effects. The study underscores the necessity for larger and longitudinal studies to accurately measure and validate the effectiveness of quantitative methods in clinical BC management.

Reproductive risk factors for breast cancer and association with novel breast density measurements among Hispanic, Black, and White women

PURPOSE

There are differences in the distributions of breast cancer incidence and risk factors by race and ethnicity. Given the strong association between breast density and breast cancer, it is of interest describe racial and ethnic variation in the determinants of breast density.

METHODS

We characterized racial and ethnic variation in reproductive history and several measures of breast density for Hispanic (n = 286), non-Hispanic Black (n = 255), and non-Hispanic White (n = 1694) women imaged at a single hospital. We quantified associations between reproductive factors and percent volumetric density (PVD), dense volume (DV), non-dense volume (NDV), and a novel measure of pixel intensity variation (V) using multivariable-adjusted linear regression, and tested for statistical heterogeneity by race and ethnicity.

RESULTS

Reproductive factors most strongly associated with breast density were age at menarche, parity, and oral contraceptive use. Variation by race and ethnicity was most evident for the associations between reproductive factors and NDV (minimum p-heterogeneity:0.008) and V (minimum p-heterogeneity:0.004) and least evident for PVD (minimum p-heterogeneity:0.042) and DV (minimum p-heterogeneity:0.041).

CONCLUSION

Reproductive choices, particularly those related to childbearing and oral contraceptive use, may contribute to racial and ethnic variation in breast density.

Computer-extracted global radiomic features can predict the radiologists' first impression about the abnormality of a screening mammogram

OBJECTIVE

Radiologists can detect the gist of abnormal based on their rapid initial impression on a mammogram (ie, global gist signal [GGS]). This study explores (1) whether global radiomic (ie, computer-extracted) features can predict the GGS; and if so, (ii) what features are the most important drivers of the signals.

METHODS

The GGS of cases in two extreme conditions was considered: when observers detect a very strong gist (high-gist) and when the gist of abnormal was not/poorly perceived (low-gist). Gist signals/scores from 13 observers reading 4191 craniocaudal mammograms were collected. As gist is a noisy signal, the gist scores from all observers were averaged and assigned to each image. The high-gist and low-gist categories contained all images in the fourth and first quartiles, respectively. One hundred thirty handcrafted global radiomic features (GRFs) per mammogram were extracted and utilized to construct eight separate machine learning random forest classifiers (All, Normal, Cancer, Prior-1, Prior-2, Missed, Prior-Visible, and Prior-Invisible) for characterizing high-gist from low-gist images. The models were trained and validated using the 10-fold cross-validation approach. The models' performances were evaluated by the area under receiver operating characteristic curve (AUC). Important features for each model were identified through a scree test.

RESULTS

The Prior-Visible model achieved the highest AUC of 0.84 followed by the Prior-Invisible (0.83), Normal (0.82), Prior-1 (0.81), All (0.79), Prior-2 (0.77), Missed (0.75), and Cancer model (0.69). Cluster shade, standard deviation, skewness, kurtosis, and range were identified to be the most important features.

CONCLUSIONS

Our findings suggest that GRFs can accurately classify high- from low-gist images.

ADVANCES IN KNOWLEDGE

Global mammographic radiomic features can accurately predict high- from low-gist images with five features identified to be valuable in describing high-gist images. These are critical in providing better understanding of the mammographic image characteristics that drive the strength of the GGSs which could be exploited to advance breast cancer (BC) screening and risk prediction, enabling early detection and treatment of BC thereby further reducing BC-related deaths.

Effect of testosterone therapy on breast tissue composition and mammographic breast density in trans masculine individuals

Objective

Determine the association between TT and breast tissue composition and breast tissue density in trans masculine individuals (TMIs).

Design

This is a cross-sectional study.

Setting

TMIs (n=444) underwent chest-contouring surgeries to treat their gender dysphoria between 2013 and 2019 at an urban medical center.

Participants

Of the 444 TMIs, 425 had pathology images analyzed by our deep-learning algorithm to extract breast tissue composition. A subset of 42/444 TMIs had mammography prior to surgery; mammography files were available for 25/42 TMIs and analyzed using a breast density software, LIBRA.

Main Outcomes and Measures

The first outcome was the association of duration of TT and breast tissue composition assessed by pathologists (categories of lobular atrophy and stromal composition) or by our algorithm (% epithelium, % fibrous stroma, and % fat). The second outcome is the association of TT and breast density as assessed by a radiologist (categorical variable) or by LIBRA (percent density, absolute dense area, and absolute non-dense area).

Results

Length of TT was associated with increasing degrees of lobular atrophy ( p <0.001) but not fibrous content ( p =0.821) when assessed by the pathologists. Every six months of TT was associated with decreased amounts of both epithelium (exp(β)=0.97, 95% CI 0.95-0.98, adj p =0.005) and stroma (exp(β)=0.99, 95% CI 0.98-1.00, adj p =0.051), but not fat (exp(β)=1.01, 95%CI 0.98-1.05, p =0.394) in fully adjusted models. There was no association between TT and radiologist's breast density assessment ( p =0.575) or LIBRA measurements ( p >0.05).

Conclusions

TT decreases breast epithelium and fibrous stroma, thus potentially reducing the breast cancer risk of TMIs. Further studies are warranted to elucidate the effect of TT on breast density and breast cancer risk.

Summary Box

Very little is known about the effect of gender-affirming testosterone therapy on cancer risks, such as breast cancer.Epidemiological studies had different conclusions about the association between testosterone and breast cancer in cisgender women (positive association) and trans masculine individuals (inverse association).More laboratory-based research are needed to understand the effect of testosterone on breast cancer risk in the understudied trans masculine population.Our study provides quantitative histological evidence to support prior epidemiological reports that testosterone may reduce breast cancer risk in trans masculine individuals.

Pectoral muscle removal in mammogram images: A novel approach for improved accuracy and efficiency

PURPOSE

Accurate pectoral muscle removal is critical in mammographic breast density estimation and many other computer-aided algorithms. We propose a novel approach to remove pectoral muscles form mediolateral oblique (MLO) view mammograms and compare accuracy and computational efficiency with existing method (Libra).

METHODS

A pectoral muscle identification pipeline was developed. The image is first binarized to enhance contrast and then the Canny algorithm was applied for edge detection. Robust interpolation is used to smooth out the pectoral muscle region. Accuracy and computational speed of pectoral muscle identification was assessed using 951 women (1,902 MLO mammograms) from the Joanne Knight Breast Health Cohort at Washington University School of Medicine.

RESULTS

Our proposed algorithm exhibits lower mean error of 12.22% in comparison to Libra's estimated error of 20.44%. This 40% gain in accuracy was statistically significant (p < 0.001). The computational time for the proposed algorithm is 5.4 times faster when compared to Libra (5.1 s for proposed vs. 27.7 s for Libra per mammogram).

CONCLUSION

We present a novel approach for pectoral muscle removal in mammogram images that demonstrates significant improvement in accuracy and efficiency compared to existing method. Our findings have important implications for the development of computer-aided systems and other automated tools in this field.

Risk Analysis of Breast Cancer by Using Bilateral Mammographic Density Differences: A Case-Control Study

The identification of risk factors helps radiologists assess the risk of breast cancer. Quantitative factors such as age and mammographic density are established risk factors for breast cancer. Asymmetric breast findings are frequently encountered during diagnostic mammography. The asymmetric area may indicate a developing mass in the early stage, causing a difference in mammographic density between the left and right sides. Therefore, this paper aims to propose a quantitative parameter named bilateral mammographic density difference (BMDD) for the quantification of breast asymmetry and to verify BMDD as a risk factor for breast cancer. To quantitatively evaluate breast asymmetry, we developed a semi-automatic method to estimate mammographic densities and calculate BMDD as the absolute difference between the left and right mammographic densities. And then, a retrospective case-control study, covering the period from July 2006 to October 2014, was conducted to analyse breast cancer risk in association with BMDD. The study included 364 women diagnosed with breast cancer and 364 matched control patients. As a result, a significant difference in BMDD was found between cases and controls (P < 0.001) and the case-control study demonstrated that women with BMDD > 10% had a 2.4-fold higher risk of breast cancer (odds ratio, 2.4; 95% confidence interval, 1.3-4.5) than women with BMDD ≤ 10%. In addition, we also demonstrated the positive association between BMDD and breast cancer risk among the subgroups with different ages and the Breast Imaging Reporting and Data System (BI-RADS) mammographic density categories. This study demonstrated that BMDD could be a potential risk factor for breast cancer.

Impact of breast density on diagnostic accuracy in digital breast tomosynthesis versus digital mammography: results from a European screening trial

BACKGROUND

The diagnostic accuracy of digital breast tomosynthesis (DBT) and digital mammography (DM) in breast cancer screening may vary per breast density subgroup. The purpose of this study was to evaluate which women, based on automatically assessed breast density subgroups, have the greatest benefit of DBT compared with DM in the prospective Malmö Breast Tomosynthesis Screening Trial.

MATERIALS AND METHODS

The prospective European, Malmö Breast Tomosynthesis Screening Trial (n = 14,848, Jan. 27, 2010-Feb. 13, 2015) compared one-view DBT and two-view DM, with consensus meeting before recall. Breast density was assessed in this secondary analysis with the automatic software Laboratory for Individualized Breast Radiodensity Assessment. DBT and DM's diagnostic accuracies were compared by breast density quintiles of breast percent density (PD) and absolute dense area (DA) with confidence intervals (CI) and McNemar's test. The association between breast density and cancer detection was analyzed with logistic regression, adjusted for ages < 55 and ≥ 55 years and previous screening participation.

RESULTS

In total, 14,730 women (median age: 58 years; inter-quartile range = 16) were included in the analysis. Sensitivity was higher and specificity lower for DBT compared with DM for all density subgroups. The highest breast PD quintile showed the largest difference in sensitivity and specificity at 81.1% (95% CI 65.8-90.5) versus 43.2% (95% CI 28.7-59.1), p < .001 and 95.5% (95% CI 94.7-96.2) versus 97.2% (95% CI 96.6-97.8), p < 0.001, respectively. Breast PD quintile was also positively associated with cancer detected via DBT at odds ratio 1.24 (95% CI 1.09-1.42, p = 0.001).

CONCLUSION

Women with the highest breast density had the greatest benefit from digital breast tomosynthesis compared with digital mammography with increased sensitivity at the cost of slightly lower specificity. These results may influence digital breast tomosynthesis's use in an individualized screening program stratified by, for instance, breast density.

TRIAL REGISTRATION

Trial registration at https://www.

CLINICALTRIALS

gov : NCT01091545, registered March 24, 2010.

Deep Learning for Fully Automatic Tumor Segmentation on Serially Acquired Dynamic Contrast-Enhanced MRI Images of Triple-Negative Breast Cancer

Accurate tumor segmentation is required for quantitative image analyses, which are increasingly used for evaluation of tumors. We developed a fully automated and high-performance segmentation model of triple-negative breast cancer using a self-configurable deep learning framework and a large set of dynamic contrast-enhanced MRI images acquired serially over the patients' treatment course. Among all models, the top-performing one that was trained with the images across different time points of a treatment course yielded a Dice similarity coefficient of 93% and a sensitivity of 96% on baseline images. The top-performing model also produced accurate tumor size measurements, which is valuable for practical clinical applications.

Epidemiology and Risk Factors for Breast Cancer: 21st Century Advances, Gaps to Address through Interdisciplinary Science

Research methods to study risk factors and prevention of breast cancer have evolved rapidly. We focus on advances from epidemiologic studies reported over the past two decades addressing scientific discoveries, as well as their clinical and public health translation for breast cancer risk reduction. In addition to reviewing methodology advances such as widespread assessment of mammographic density and Mendelian randomization, we summarize the recent evidence with a focus on the timing of exposure and windows of susceptibility. We summarize the implications of the new evidence for application in risk stratification models and clinical translation to focus prevention-maximizing benefits and minimizing harm. We conclude our review identifying research gaps. These include: pathways for the inverse association of vegetable intake and estrogen receptor (ER)-ve tumors, prepubertal and adolescent diet and risk, early life adiposity reducing lifelong risk, and gaps from changes in habits (e.g., vaping, binge drinking), and environmental exposures.

Association and Prediction Utilizing Craniocaudal and Mediolateral Oblique View Digital Mammography and Long-Term Breast Cancer Risk

Mammographic percentage of volumetric density is an important risk factor for breast cancer. Epidemiology studies historically used film images often limited to craniocaudal (CC) views to estimate area-based breast density. More recent studies using digital mammography images typically use the averaged density between craniocaudal (CC) and mediolateral oblique (MLO) view mammography for 5- and 10-year risk prediction. The performance in using either and both mammogram views has not been well-investigated. We use 3,804 full-field digital mammograms from the Joanne Knight Breast Health Cohort (294 incident cases and 657 controls), to quantity the association between volumetric percentage of density extracted from either and both mammography views and to assess the 5 and 10-year breast cancer risk prediction performance. Our results show that the association between percent volumetric density from CC, MLO, and the average between the two, retain essentially the same association with breast cancer risk. The 5- and 10-year risk prediction also shows similar prediction accuracy. Thus, one view is sufficient to assess association and predict future risk of breast cancer over a 5 or 10-year interval.

PREVENTION RELEVANCE

Expanding use of digital mammography and repeated screening provides opportunities for risk assessment. To use these images for risk estimates and guide risk management in real time requires efficient processing. Evaluating the contribution of different views to prediction performance can guide future applications for risk management in routine care.

Examination of fully automated mammographic density measures using LIBRA and breast cancer risk in a cohort of 21,000 non-Hispanic white women

BACKGROUND

Breast density is strongly associated with breast cancer risk. Fully automated quantitative density assessment methods have recently been developed that could facilitate large-scale studies, although data on associations with long-term breast cancer risk are limited. We examined LIBRA assessments and breast cancer risk and compared results to prior assessments using Cumulus, an established computer-assisted method requiring manual thresholding.

METHODS

We conducted a cohort study among 21,150 non-Hispanic white female participants of the Research Program in Genes, Environment and Health of Kaiser Permanente Northern California who were 40-74 years at enrollment, followed for up to 10 years, and had archived processed screening mammograms acquired on Hologic or General Electric full-field digital mammography (FFDM) machines and prior Cumulus density assessments available for analysis. Dense area (DA), non-dense area (NDA), and percent density (PD) were assessed using LIBRA software. Cox regression was used to estimate hazard ratios (HRs) for breast cancer associated with DA, NDA and PD modeled continuously in standard deviation (SD) increments, adjusting for age, mammogram year, body mass index, parity, first-degree family history of breast cancer, and menopausal hormone use. We also examined differences by machine type and breast view.

RESULTS

The adjusted HRs for breast cancer associated with each SD increment of DA, NDA and PD were 1.36 (95% confidence interval, 1.18-1.57), 0.85 (0.77-0.93) and 1.44 (1.26-1.66) for LIBRA and 1.44 (1.33-1.55), 0.81 (0.74-0.89) and 1.54 (1.34-1.77) for Cumulus, respectively. LIBRA results were generally similar by machine type and breast view, although associations were strongest for Hologic machines and mediolateral oblique views. Results were also similar during the first 2 years, 2-5 years and 5-10 years after the baseline mammogram.

CONCLUSION

Associations with breast cancer risk were generally similar for LIBRA and Cumulus density measures and were sustained for up to 10 years. These findings support the suitability of fully automated LIBRA assessments on processed FFDM images for large-scale research on breast density and cancer risk.

Longitudinal Analysis of Change in Mammographic Density in Each Breast and Its Association With Breast Cancer Risk

Importance

Although breast density is an established risk factor for breast cancer, longitudinal changes in breast density have not been extensively studied to determine whether this factor is associated with breast cancer risk.

Objective

To prospectively evaluate the association between change in mammographic density in each breast over time and risk of subsequent breast cancer.

Design, Setting, and Participants

This nested case-control cohort study was sampled from the Joanne Knight Breast Health Cohort of 10 481 women free from cancer at entry and observed from November 3, 2008, to October 31, 2020, with routine screening mammograms every 1 to 2 years, providing a measure of breast density. Breast cancer screening was provided for a diverse population of women in the St Louis region. A total of 289 case patients with pathology-confirmed breast cancer were identified, and approximately 2 control participants were sampled for each case according to age at entry and year of enrollment, yielding 658 controls with a total number of 8710 craniocaudal-view mammograms for analysis.

Exposures

Exposures included screening mammograms with volumetric percentage of density, change in volumetric breast density over time, and breast biopsy pathology-confirmed cancer. Breast cancer risk factors were collected via questionnaire at enrollment.

Main Outcomes and Measures

Longitudinal changes over time in each woman's volumetric breast density by case and control status.

Results

The mean (SD) age of the 947 participants was 56.67 (8.71) years at entry; 141 were Black (14.9%), 763 were White (80.6%), 20 were of other race or ethnicity (2.1%), and 23 did not report this information (2.4%). The mean (SD) interval was 2.0 (1.5) years from last mammogram to date of subsequent breast cancer diagnosis (10th percentile, 1.0 year; 90th percentile, 3.9 years). Breast density decreased over time in both cases and controls. However, there was a significantly slower decrease in rate of decline in density in the breast that developed breast cancer compared with the decline in controls (estimate = 0.027; 95% CI, 0.001-0.053; P = .04).

Conclusions and Relevance

This study found that the rate of change in breast density was associated with the risk of subsequent breast cancer. Incorporation of longitudinal changes into existing models could optimize risk stratification and guide more personalized risk management.

Effects of bariatric surgery on breast density in adult obese women: systematic review and meta-analysis

Introduction

Bariatric surgery is one of the most effective methods for treating obesity. It can effectively reduce body weight and reduce the incidence of obesity-related breast cancer. However, there are different conclusions about how bariatric surgery changes breast density. The purpose of this study was to clarify the changes in breast density from before to after bariatric surgery.

Methods

The relevant literature was searched through PubMed and Embase to screen for studies. Meta-analysis was used to clarify the changes in breast density from before to after bariatric surgery.

Results

A total of seven studies were included in this systematic review and meta-analysis, including a total of 535 people. The average body mass index decreased from 45.3 kg/m2 before surgery to 34.4 kg/m2 after surgery. By the Breast Imaging Reporting and Data System score, the proportion of grade A breast density from before to after bariatric surgery decreased by 3.83% (183 vs. 176), grade B (248 vs. 263) increased by 6.05%, grade C (94 vs. 89) decreased by 5.32%, and grade D (1 vs. 4) increased by 300%. There was no significant change in breast density from before to after bariatric surgery (OR=1.27, 95% confidence interval (CI) [0.74, 2.20], P=0.38). By the Volpara density grade score, postoperative volumetric breast density increased (standardized mean difference = -0.68, 95% CI [-1.08, -0.27], P = 0.001).

Discussions

Breast density increased significantly after bariatric surgery, but this depended on the method of detecting breast density. Further randomized controlled studies are needed to validate our conclusions.

Using Radiomics-Based Machine Learning to Create Targeted Test Sets to Improve Specific Mammography Reader Cohort Performance: A Feasibility Study

Mammography interpretation is challenging with high error rates. This study aims to reduce the errors in mammography reading by mapping diagnostic errors against global mammographic characteristics using a radiomics-based machine learning approach. A total of 36 radiologists from cohort A (n = 20) and cohort B (n = 16) read 60 high-density mammographic cases. Radiomic features were extracted from three regions of interest (ROIs), and random forest models were trained to predict diagnostic errors for each cohort. Performance was evaluated using sensitivity, specificity, accuracy, and AUC. The impact of ROI placement and normalization on prediction was investigated. Our approach successfully predicted both the false positive and false negative errors of both cohorts but did not consistently predict location errors. The errors produced by radiologists from cohort B were less predictable compared to those in cohort A. The performance of the models did not show significant improvement after feature normalization, despite the mammograms being produced by different vendors. Our novel radiomics-based machine learning pipeline focusing on global radiomic features could predict false positive and false negative errors. The proposed method can be used to develop group-tailored mammographic educational strategies to help improve future mammography reader performance.

Genome-Wide Association Study of Breast Density among Women of African Ancestry

Breast density, the amount of fibroglandular versus fatty tissue in the breast, is a strong breast cancer risk factor. Understanding genetic factors associated with breast density may help in clarifying mechanisms by which breast density increases cancer risk. To date, 50 genetic loci have been associated with breast density, however, these studies were performed among predominantly European ancestry populations. We utilized a cohort of women aged 40-85 years who underwent screening mammography and had genetic information available from the Penn Medicine BioBank to conduct a Genome-Wide Association Study (GWAS) of breast density among 1323 women of African ancestry. For each mammogram, the publicly available "LIBRA" software was used to quantify dense area and area percent density. We identified 34 significant loci associated with dense area and area percent density, with the strongest signals in GACAT3, CTNNA3, HSD17B6, UGDH, TAAR8, ARHGAP10, BOD1L2, and NR3C2. There was significant overlap between previously identified breast cancer SNPs and SNPs identified as associated with breast density. Our results highlight the importance of breast density GWAS among diverse populations, including African ancestry populations. They may provide novel insights into genetic factors associated with breast density and help in elucidating mechanisms by which density increases breast cancer risk.

The effect of variable labels on deep learning models trained to predict breast density

Purpose. High breast density is associated with reduced efficacy of mammographic screening and increased risk of developing breast cancer. Accurate and reliable automated density estimates can be used for direct risk prediction and passing density related information to further predictive models. Expert reader assessments of density show a strong relationship to cancer risk but also inter-reader variation. The effect of label variability on model performance is important when considering how to utilise automated methods for both research and clinical purposes.Methods. We utilise subsets of images with density labels from the same 13 readers and 12 reader pairs, and train a deep transfer learning model which is used to assess how label variability affects the mapping from representation to prediction. We then create two end-to-end models: one that is trained on averaged labels across the reader pairs and the second that is trained using individual reader scores, with a novel alteration to the objective function. The combination of these two end-to-end models allows us to investigate the effect of label variability on the model representation formed.Results. We show that the trained mappings from representations to labels are altered considerably by the variability of reader scores. Training on labels with distribution variation removed causes the Spearman rank correlation coefficients to rise from 0.751 ± 0.002 to either 0.815 ± 0.026 when averaging across readers or 0.844 ± 0.002 when averaging across images. However, when we train different models to investigate the representation effect we see little difference, with Spearman rank correlation coefficients of 0.846 ± 0.006 and 0.850 ± 0.006 showing no statistically significant difference in the quality of the model representation with regard to density prediction.Conclusions. We show that the mapping between representation and mammographic density prediction is significantly affected by label variability. However, the effect of the label variability on the model representation is limited.

Changes in mammographic density and risk of breast cancer among a diverse cohort of women undergoing mammography screening

PURPOSE

Mammographic density (MD) is a strong breast cancer risk factor. MD may change over time, with potential implications for breast cancer risk. Few studies have assessed associations between MD change and breast cancer in racially diverse populations. We investigated the relationships between MD and MD change over time and breast cancer risk in a large, diverse screening cohort.

MATERIALS AND METHODS

We retrospectively analyzed data from 8462 women who underwent ≥ 2 screening mammograms from Sept. 2010 to Jan. 2015 (N = 20,766 exams); 185 breast cancers were diagnosed 1-7 years after screening. Breast percent density (PD) and dense area (DA) were estimated from raw digital mammograms (Hologic Inc.) using LIBRA (v1.0.4). For each MD measure, we modeled breast density change between two sequential visits as a function of demographic and risk covariates. We used Cox regression to examine whether varying degrees of breast density change were associated with breast cancer risk, accounting for multiple exams per woman.

RESULTS

PD at any screen was significantly associated with breast cancer risk (hazard ratio (HR) for PD = 1.03 (95% CI [1.01, 1.05], p < 0.0005), but neither change in breast density nor more extreme than expected changes in breast density were associated with breast cancer risk. We found no evidence of differences in density change or breast cancer risk due to density change by race. Results using DA were essentially identical.

CONCLUSIONS

Using a large racially diverse cohort, we found no evidence of association between short-term change in MD and risk of breast cancer, suggesting that short-term MD change is not a strong predictor for risk.

Automatic assessment of mammographic density using a deep transfer learning method

Purpose

Mammographic breast density is one of the strongest risk factors for cancer. Density assessed by radiologists using visual analogue scales has been shown to provide better risk predictions than other methods. Our purpose is to build automated models using deep learning and train on radiologist scores to make accurate and consistent predictions.

Approach

We used a dataset of almost 160,000 mammograms, each with two independent density scores made by expert medical practitioners. We used two pretrained deep networks and adapted them to produce feature vectors, which were then used for both linear and nonlinear regression to make density predictions. We also simulated an "optimal method," which allowed us to compare the quality of our results with a simulated upper bound on performance.

Results

Our deep learning method produced estimates with a root mean squared error (RMSE) of 8.79 ± 0.21 . The model estimates of cancer risk perform at a similar level to human experts, within uncertainty bounds. We made comparisons between different model variants and demonstrated the high level of consistency of the model predictions. Our modeled "optimal method" produced image predictions with a RMSE of between 7.98 and 8.90 for cranial caudal images.

Conclusion

We demonstrated a deep learning framework based upon a transfer learning approach to make density estimates based on radiologists' visual scores. Our approach requires modest computational resources and has the potential to be trained with limited quantities of data.

Artificial intelligence in breast cancer imaging: risk stratification, lesion detection and classification, treatment planning and prognosis-a narrative review

The advent of artificial intelligence (AI) represents a real game changer in today's landscape of breast cancer imaging. Several innovative AI-based tools have been developed and validated in recent years that promise to accelerate the goal of real patient-tailored management. Numerous studies confirm that proper integration of AI into existing clinical workflows could bring significant benefits to women, radiologists, and healthcare systems. The AI-based approach has proved particularly useful for developing new risk prediction models that integrate multi-data streams for planning individualized screening protocols. Furthermore, AI models could help radiologists in the pre-screening and lesion detection phase, increasing diagnostic accuracy, while reducing workload and complications related to overdiagnosis. Radiomics and radiogenomics approaches could extrapolate the so-called imaging signature of the tumor to plan a targeted treatment. The main challenges to the development of AI tools are the huge amounts of high-quality data required to train and validate these models and the need for a multidisciplinary team with solid machine-learning skills. The purpose of this article is to present a summary of the most important AI applications in breast cancer imaging, analyzing possible challenges and new perspectives related to the widespread adoption of these new tools.

Deep learning of longitudinal mammogram examinations for breast cancer risk prediction

Information in digital mammogram images has been shown to be associated with the risk of developing breast cancer. Longitudinal breast cancer screening mammogram examinations may carry spatiotemporal information that can enhance breast cancer risk prediction. No deep learning models have been designed to capture such spatiotemporal information over multiple examinations to predict the risk. In this study, we propose a novel deep learning structure, LRP-NET, to capture the spatiotemporal changes of breast tissue over multiple negative/benign screening mammogram examinations to predict near-term breast cancer risk in a case-control setting. Specifically, LRP-NET is designed based on clinical knowledge to capture the imaging changes of bilateral breast tissue over four sequential mammogram examinations. We evaluate our proposed model with two ablation studies and compare it to three models/settings, including 1) a "loose" model without explicitly capturing the spatiotemporal changes over longitudinal examinations, 2) LRP-NET but using a varying number (i.e., 1 and 3) of sequential examinations, and 3) a previous model that uses only a single mammogram examination. On a case-control cohort of 200 patients, each with four examinations, our experiments on a total of 3200 images show that the LRP-NET model outperforms the compared models/settings.

Comparison between two packages for pectoral muscle removal on mammographic images

Background

Pectoral muscle removal is a fundamental preliminary step in computer-aided diagnosis systems for full-field digital mammography (FFDM). Currently, two open-source publicly available packages (LIBRA and OpenBreast) provide algorithms for pectoral muscle removal within Matlab environment.

Purpose

To compare performance of the two packages on a single database of FFDM images.

Methods

Only mediolateral oblique (MLO) FFDM was considered because of large presence of pectoral muscle on this type of projection. For obtaining ground truth, pectoral muscle has been manually segmented by two radiologists in consensus. Both LIBRA’s and OpenBreast’s removal performance with respect to ground truth were compared using Dice similarity coefficient and Cohen-kappa reliability coefficient; Wilcoxon signed-rank test has been used for assessing differences in performances; Kruskal–Wallis test has been used to verify possible dependence of the performance from the breast density or image laterality.

Results

FFDMs from 168 consecutive women at our institution have been included in the study. Both LIBRA’s Dice-index and Cohen-kappa were significantly higher than OpenBreast (Wilcoxon signed-rank test P < 0.05). No dependence on breast density or laterality has been found (Kruskal–Wallis test P > 0.05). Conclusion: Libra has a better performance than OpenBreast in pectoral muscle delineation so that, although our study has not a direct clinical application, these results are useful in the choice of packages for the development of complex systems for computer-aided breast evaluation.

Area-based breast percentage density estimation in mammograms using weight-adaptive multitask learning

Breast density, which is a measure of the relative amount of fibroglandular tissue within the breast area, is one of the most important breast cancer risk factors. Accurate segmentation of fibroglandular tissues and breast area is crucial for computing the breast density. Semiautomatic and fully automatic computer-aided design tools have been developed to estimate the percentage of breast density in mammograms. However, the available approaches are usually limited to specific mammogram views and are inadequate for complete delineation of the pectoral muscle. These tools also perform poorly in cases of data variability and often require an experienced radiologist to adjust the segmentation threshold for fibroglandular tissue within the breast area. This study proposes a new deep learning architecture that automatically estimates the area-based breast percentage density from mammograms using a weight-adaptive multitask learning approach. The proposed approach simultaneously segments the breast and dense tissues and further estimates the breast percentage density. We evaluate the performance of the proposed model in both segmentation and density estimation on an independent evaluation set of 7500 craniocaudal and mediolateral oblique-view mammograms from Kuopio University Hospital, Finland. The proposed multitask segmentation approach outperforms and achieves average relative improvements of 2.88% and 9.78% in terms of F-score compared to the multitask U-net and a fully convolutional neural network, respectively. The estimated breast density values using our approach strongly correlate with radiologists' assessments with a Pearson's correlation of [Formula: see text] (95% confidence interval [0.89, 0.91]). We conclude that our approach greatly improves the segmentation accuracy of the breast area and dense tissues; thus, it can play a vital role in accurately computing the breast density. Our density estimation model considerably reduces the time and effort needed to estimate density values from mammograms by radiologists and therefore, decreases inter- and intra-reader variability.

A deep learning framework to classify breast density with noisy labels regularization

BACKGROUND AND OBJECTIVE

Breast density assessed from digital mammograms is a biomarker for higher risk of developing breast cancer. Experienced radiologists assess breast density using the Breast Image and Data System (BI-RADS) categories. Supervised learning algorithms have been developed with this objective in mind, however, the performance of these algorithms depends on the quality of the ground-truth information which is usually labeled by expert readers. These labels are noisy approximations of the ground truth, as there is often intra- and inter-reader variability among labels. Thus, it is crucial to provide a reliable method to obtain digital mammograms matching BI-RADS categories. This paper presents RegL (Labels Regularizer), a methodology that includes different image pre-processes to allow both a correct breast segmentation and the enhancement of image quality through an intensity adjustment, thus allowing the use of deep learning to classify the mammograms into BI-RADS categories. The Confusion Matrix (CM) - CNN network used implements an architecture that models each radiologist's noisy label. The final methodology pipeline was determined after comparing the performance of image pre-processes combined with different DL architectures.

METHODS

A multi-center study composed of 1395 women whose mammograms were classified into the four BI-RADS categories by three experienced radiologists is presented. A total of 892 mammograms were used as the training corpus, 224 formed the validation corpus, and 279 the test corpus.

RESULTS

The combination of five networks implementing the RegL methodology achieved the best results among all the models in the test set. The ensemble model obtained an accuracy of (0.85) and a kappa index of 0.71.

CONCLUSIONS

The proposed methodology has a similar performance to the experienced radiologists in the classification of digital mammograms into BI-RADS categories. This suggests that the pre-processing steps and modelling of each radiologist's label allows for a better estimation of the unknown ground truth labels.

STATISTICAL MODELING OF GLANDULARITY FROM MAMMOGRAPHY IMAGES

This study presents a methodology for estimation of breast glandularity, which is an important factor to assess radiological risk in mammography patients. The investigation took place in an institution located at department of Antioquia-Colombia, where 200 patients participated. The models were obtained using partial least squares regression, where Dance's model was used as reference; parameters of mammography images, equipment and patient were used as predicting variables (kV, mAs, patient's weight, breast area and mean gray value of breast images). Coefficients of correlation equal to 89 and 88 were obtained for training and validation respectively in mediolateral oblique (MLO) and 84 and 89 for craniocaudal (CC). These models were used to estimate the mean glandular dose for all patients and later to obtain the institutional reference levels, 0.87 and 0.96 mGy for CC and MLO, respectively, following the recommendations of the ICRP publication No. 135. This study suggests that glandularity could be estimated with few parameters from equipment and patient.

Unsupervised domain adaptation for the segmentation of breast tissue in mammography images

BACKGROUND AND OBJECTIVE

Breast density refers to the proportion of glandular and fatty tissue in the breast and is recognized as a useful factor assessing breast cancer risk. Moreover, the segmentation of the high-density glandular tissue from mammograms can assist medical professionals visualizing and localizing areas that may require additional attention. Developing robust methods to segment breast tissues is challenging due to the variations in mammographic acquisition systems and protocols. Deep learning methods are effective in medical image segmentation but they often require large quantities of labelled data. Unsupervised domain adaptation is an area of research that employs unlabelled data to improve model performance on variations of samples derived from different sources.

METHODS

First, a U-Net architecture was used to perform segmentation of the fatty and glandular tissues with labelled data from a single acquisition device. Then, adversarial-based unsupervised domain adaptation methods were used to incorporate single unlabelled target domains, consisting of images from a different machine, into the training. Finally, the domain adaptation model was extended to include multiple unlabelled target domains by combining a reconstruction task with adversarial training.

RESULTS

The adversarial training was found to improve the generalization of the initial model on new domain data, demonstrating clearly improved segmentation of the breast tissues. For training with multiple unlabelled domains, combining a reconstruction task with adversarial training improved the stability of the training and yielded adequate segmentation results across all domains with a single model.

CONCLUSIONS

Results demonstrated the potential for adversarial-based domain adaptation with U-Net architectures for segmentation of breast tissue in mammograms coming from several devices and demonstrated that domain-adapted models could achieve a similar agreement with manual segmentations. It has also been found that combining adversarial and reconstruction-based methods can provide a simple and effective solution for training with multiple unlabelled target domains.

Algorithms and methods for computerized analysis of mammography images in breast cancer risk assessment

BACKGROUND AND OBJECTIVES

The computerized analysis of mammograms for the development of quantitative biomarkers is a growing field with applications in breast cancer risk assessment. Computerized image analysis offers the possibility of using different methods and algorithms to extract additional information from screening and diagnosis images to aid in the assessment of breast cancer risk. In this work, we review the algorithms and methods for the automated, computerized analysis of mammography images for the task mentioned, and discuss the main challenges that the development and improvement of these methods face today.

METHODS

We review the recent progress in two main branches of mammography-based risk assessment: parenchymal analysis and breast density estimation, including performance indicators of most of the studies considered. Parenchymal analysis methods are divided into feature-based methods and deep learning-based methods; breast density methods are grouped into area-based, volume-based, and breast categorization methods. Additionally, we identify the challenges that these study fields currently face.

RESULTS

Parenchymal analysis using deep learning algorithms are on the rise, with some studies showing high-performance indicators, such as an area under the receiver operating characteristic curve of up to 90. Methods for risk assessment using breast density report a wider variety of performance indicators; however, we can also identify that the approaches using deep learning methods yield high performance in each of the subdivisions considered.

CONCLUSIONS

Both breast density estimation and parenchymal analysis are promising tools for the task of breast cancer risk assessment; deep learning methods have shown performance comparable or superior to the other considered methods. All methods considered face challenges such as the lack of objective comparison between them and the lack of access to datasets from different populations.

Incorporating Robustness to Imaging Physics into Radiomic Feature Selection for Breast Cancer Risk Estimation

Digital mammography has seen an explosion in the number of radiomic features used for risk-assessment modeling. However, having more features is not necessarily beneficial, as some features may be overly sensitive to imaging physics (contrast, noise, and image sharpness). To measure the effects of imaging physics, we analyzed the feature variation across imaging acquisition settings (kV, mAs) using an anthropomorphic phantom. We also analyzed the intra-woman variation (IWV), a measure of how much a feature varies between breasts with similar parenchymal patterns-a woman's left and right breasts. From 341 features, we identified "robust" features that minimized the effects of imaging physics and IWV. We also investigated whether robust features offered better case-control classification in an independent data set of 575 images, all with an overall BI-RADS® assessment of 1 (negative) or 2 (benign); 115 images (cases) were of women who developed cancer at least one year after that screening image, matched to 460 controls. We modeled cancer occurrence via logistic regression, using cross-validated area under the receiver-operating-characteristic curve (AUC) to measure model performance. Models using features from the most-robust quartile of features yielded an AUC = 0.59, versus 0.54 for the least-robust, with p < 0.005 for the difference among the quartiles.

Radiomic features of breast parenchyma: assessing differences between FOR PROCESSING and FOR PRESENTATION digital mammography

OBJECTIVE

To assess the similarity and differences of radiomics features on full field digital mammography (FFDM) in FOR PROCESSING and FOR PRESENTATION data.

METHODS

165 consecutive women who underwent FFDM were included. Breasts have been segmented into "dense" and "non-dense" area using the software LIBRA. Segmentation of both FOR PROCESSING and FOR PRESENTATION images have been evaluated by Bland-Altman, Dice index and Cohen's kappa analysis. 74 textural features were computed: 18 features of First Order (FO), 24 features of Gray Level Co-occurrence Matrix (GLCM), 16 features of Gray Level Run Length Matrix (GLRLM) and 16 features of Gray Level Size Zone Matrix (GLSZM). Paired Wilcoxon test, Spearman's rank correlation, intraclass correlation and canonical correlation have been used. Bilateral symmetry and percent density (PD) were also evaluated.

RESULTS

Segmentation from FOR PROCESSING and FOR PRESENTATION gave very different results. Bilateral symmetry was higher when evaluated on features computed using FOR PROCESSING images. All features showed a positive Spearman's correlation coefficient and many FOR-PROCESSING features were moderately or strongly correlated to their corresponding FOR-PRESENTATION counterpart. As regards the correlation analysis between PD and textural features from FOR-PRESENTATION a moderate correlation was obtained only for Gray Level Non Uniformity from GLRLM both on "dense" and "non dense" area; as regards correlation between PD and features from FOR-PROCESSING a moderate correlation was observed only for Maximal Correlation Coefficient from GLCM both on "dense" and "non dense" area.

CONCLUSIONS

Texture features from FOR PROCESSING mammograms seem to be most suitable for assessing breast density.

Deep-LIBRA: An artificial-intelligence method for robust quantification of breast density with independent validation in breast cancer risk assessment

Breast density is an important risk factor for breast cancer that also affects the specificity and sensitivity of screening mammography. Current federal legislation mandates reporting of breast density for all women undergoing breast cancer screening. Clinically, breast density is assessed visually using the American College of Radiology Breast Imaging Reporting And Data System (BI-RADS) scale. Here, we introduce an artificial intelligence (AI) method to estimate breast density from digital mammograms. Our method leverages deep learning using two convolutional neural network architectures to accurately segment the breast area. An AI algorithm combining superpixel generation and radiomic machine learning is then applied to differentiate dense from non-dense tissue regions within the breast, from which breast density is estimated. Our method was trained and validated on a multi-racial, multi-institutional dataset of 15,661 images (4,437 women), and then tested on an independent matched case-control dataset of 6368 digital mammograms (414 cases; 1178 controls) for both breast density estimation and case-control discrimination. On the independent dataset, breast percent density (PD) estimates from Deep-LIBRA and an expert reader were strongly correlated (Spearman correlation coefficient = 0.90). Moreover, in a model adjusted for age and BMI, Deep-LIBRA yielded a higher case-control discrimination performance (area under the ROC curve, AUC = 0.612 [95% confidence interval (CI): 0.584, 0.640]) compared to four other widely-used research and commercial breast density assessment methods (AUCs = 0.528 to 0.599). Our results suggest a strong agreement of breast density estimates between Deep-LIBRA and gold-standard assessment by an expert reader, as well as improved performance in breast cancer risk assessment over state-of-the-art open-source and commercial methods.

Fully Automated Volumetric Breast Density Estimation from Digital Breast Tomosynthesis

Background While digital breast tomosynthesis (DBT) is rapidly replacing digital mammography (DM) in breast cancer screening, the potential of DBT density measures for breast cancer risk assessment remains largely unexplored. Purpose To compare associations of breast density estimates from DBT and DM with breast cancer. Materials and Methods This retrospective case-control study used contralateral DM/DBT studies from women with unilateral breast cancer and age- and ethnicity-matched controls (September 19, 2011-January 6, 2015). Volumetric percent density (VPD%) was estimated from DBT using previously validated software. For comparison, the publicly available Laboratory for Individualized Breast Radiodensity Assessment software package, or LIBRA, was used to estimate area-based percent density (APD%) from raw and processed DM images. The commercial Quantra and Volpara software packages were applied to raw DM images to estimate VPD% with use of physics-based models. Density measures were compared by using Spearman correlation coefficients (r), and conditional logistic regression was performed to examine density associations (odds ratios [OR]) with breast cancer, adjusting for age and body mass index. Results A total of 132 women diagnosed with breast cancer (mean age ± standard deviation [SD], 60 years ± 11) and 528 controls (mean age, 60 years ± 11) were included. Moderate correlations between DBT and DM density measures (r = 0.32-0.75; all P < .001) were observed. Volumetric density estimates calculated from DBT (OR, 2.3 [95% CI: 1.6, 3.4] per SD for VPD%_DBT) were more strongly associated with breast cancer than DM-derived density for both APD% (OR, 1.3 [95% CI: 0.9, 1.9] [P < .001] and 1.7 [95% CI: 1.2, 2.3] [P = .004] per SD for LIBRA raw and processed data, respectively) and VPD% (OR, 1.6 [95% CI: 1.1, 2.4] [P = .01] and 1.7 [95% CI: 1.2, 2.6] [P = .04] per SD for Volpara and Quantra, respectively). Conclusion The associations between quantitative breast density estimates and breast cancer risk are stronger for digital breast tomosynthesis compared with digital mammography. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Yaffe in this issue.

Deep Learning Pre-training Strategy for Mammogram Image Classification: an Evaluation Study

In this work, we assess how pre-training strategy affects deep learning performance for the task of distinguishing false-recall from malignancy and normal (benign) findings in digital mammography images. A cohort of 1303 breast cancer screening patients (4935 digital mammogram images in total) was retrospectively analyzed as the target dataset for this study. We assessed six different convolutional neural network model structures utilizing four different imaging datasets (total > 1.4 million images (including ImageNet); medical images different in terms of scale, modality, organ, and source) for pre-training on six classification tasks to assess how the performance of CNN models varies based on training strategy. Representative pre-training strategies included transfer learning with medical and non-medical datasets, layer freezing, varied network structure, and multi-view input for both binary and triple-class classification of mammogram images. The area under the receiver operating characteristic curve (AUC) was used as the model performance metric. The best performing model out of all experimental settings was an AlexNet model incrementally pre-trained on ImageNet and a large Breast Density dataset. The AUC for the six classification tasks using this model ranged from 0.68 to 0.77. In the case of distinguishing recalled-benign mammograms from others, four out of five pre-training strategies tested produced significant performance differences from the baseline model. This study suggests that pre-training strategy influences significant performance differences, especially in the case of distinguishing recalled- benign from malignant and benign screening patients.

Clinicopathologic breast cancer characteristics: predictions using global textural features of the ipsilateral breast mammogram

Radiomic features from mammograms have been shown to predict breast cancer (BC) risk; however, their contribution to BC characteristics has not yet been explored. This study included 184 women with BC between January 2012 and April 2017. A set of 33 global radiomic features were extracted from the ipsilateral breast mammogram. Associations between radiomic features and BC characteristics were investigated by univariate logistic regression analysis, and receiver-operating characteristic curve analysis was employed to evaluate the predictive performance of radiomic features. Histogram-based features (mean, 70th percentile, and 30th percentile) weakly differentiated progesterone status and tumor size (AUC range: 0.627-0.652, p ≤ 0.007). One gray level run length matrix (GLRLM)-based feature achieved an AUC of 0.68 in discriminating lymph-node status, and the fractal dimension achieved an AUC of 0.65 in predicting tumor size. After stratifying by age at BC diagnosis and baseline percent density (PD), the average predictive performance of the abovementioned features improved from 0.652 to 0.707 for baseline PD adjustment, and from 0.652 to 0.674 for age at BC diagnosis. Higher predictive performances were found for GLRLM-based features in predicting lymph-node status among younger women with high baseline PD (AUC range: 0.710-0.863), and for fractal features in predicting tumor size among patients with low PD (AUC: 0.704). Global radiomic features from the ipsilateral breast mammogram can predict lymph-node status and tumor size among certain categories of women and should be considered as a non-invasive tool for clinical decision-making in BC-affected women and for forecasting disease progression.

Breast glandularity and mean glandular dose assessment using a deep learning framework: Virtual patients study

PURPOSE

Breast dosimetry in mammography is an important aspect of radioprotection since women are exposed periodically to ionizing radiation due to breast cancer screening programs. Mean glandular dose (MGD) is the standard quantity employed for the establishment of dose reference levels in retrospective population studies. However, MGD calculations requires breast glandularity estimation. This work proposes a deep learning framework for volume glandular fraction (VGF) estimations based on mammography images, which in turn are converted to glandularity values for MGD calculations.

METHODS

208 virtual breast phantoms were generated and compressed computationally. The mammography images were obtained with Monte Carlo simulations (MC-GPU code) and a ray-tracing algorithm was employed for labeling the training data. The architectures of the neural networks are based on the XNet and multilayer perceptron, adapted for each task. The network predictions were compared with the ground truth using the coefficient of determination (r²).

RESULTS

The results have shown a good agreement for inner breast segmentation (r² = 0.999), breast volume prediction (r² = 0.982) and VGF prediction (r² = 0.935). Moreover, the DgN coefficients using the predicted VGF for the virtual population differ on average 1.3% from the ground truth values. Afterwards with the obtained DgN coefficients, the MGD values were estimated from exposure factors extracted from the DICOM header of a clinical cohort, with median(75 percentile) values of 1.91(2.45) mGy.

CONCLUSION

We successfully implemented a deep learning framework for VGF and MGD calculations for virtual breast phantoms.

Density map and fuzzy classification for breast density by using BI-RADS

Mammographic density (MD) is conformed by a different percentage of stromal, epithelial, and adipose tissue within the breast. One of the most critical findings in mammographic patterns for establishing a diagnosis of breast cancer is high breast tissue density. There is a wide variety of works focused on the study and automatic calculation of general breast density; however, they do not provide more detailed information about the changes that may occur within the breast tissue. This work proposes to generate a breast density map based on a texture analysis to identify the internal composition and distribution of the breast tissue through the diffuse division technique of the different densities inside the breast. Therefore, it is possible to obtain a density map associated with the breast that allows us to distinguish and quantify the different types of breast densities and their distribution according to the Breast Imaging Reporting and Data System (BI-RADS Breast Density Category). The proposed methodology was tested with mammograms from the BCDR and InBreast databases, demonstrating consistency in results and reaching an accuracy of 84.2% and 81.3%, respectively. Finally, the information obtained from the density map and its analysis could be a support tool for the specialist physician to monitor changes in breast density over time, since the fuzzy classification carried out allows quantifying the degree of membership in the BI-RADS breast density classes.

Lifestyle, Behavioral, and Dietary Risk Factors in Relation to Mammographic Breast Density in Women at High Risk for Breast Cancer

BACKGROUND

Women at high risk for breast cancer due to genetics or risk factor profiles are counseled to adopt lifestyle, behavioral, and dietary changes to help reduce their risk. These recommendations are based on studies of women at average risk, so their effectiveness in high-risk women is unclear.

METHODS

We evaluated the impact of physical activity, smoking, alcohol consumption, and intake of folate and carotenoids on mammographic breast density-a proxy for breast cancer risk-among 387 high-risk women. Exposures were self-reported on questionnaires. Breast dense area, nondense area, and percent dense area were measured from screening mammograms with Library for Breast Radiodensity Assessment software. Cross-sectional associations were estimated with multivariable quantile regression models.

RESULTS

After adjusting for age, adiposity, reproductive history, and use of postmenopausal hormones, no breast density measure was associated with physical activity level, smoking status, alcohol consumption, or estimated intake of folate, alpha-carotene, beta-carotene, lutein/zeaxanthin, and beta-cryptoxanthin. Lycopene intake was associated with lower dense area when comparing the highest and lowest intake categories (adjusted difference in median = -14 cm², 95% confidence interval: -29 to 1.3 cm²). This association may be explained by incomplete adjustment for adiposity.

CONCLUSIONS

Recommended lifestyle, behavioral, and dietary changes to mitigate personal risk of breast cancer do not substantially impact mammographic breast density measures.

IMPACT

Alternative strategies, such as increased uptake of chemoprevention, may better serve risk reduction efforts in women at high risk for breast cancer.

A computer-aided approach for automatic detection of breast masses in digital mammogram via spectral clustering and support vector machine

Breast cancer continues to be a widespread health concern all over the world. Mammography is an important method in the early detection of breast abnormalities. In recent years, using an automatic Computer-Aided Detection (CAD) system based on image processing techniques has been a more reliable interpretation in the illustration of breast distortion. In this study, a fully process-integrated approach with developing a CAD system is presented for the detection of breast masses based on texture description, spectral clustering, and Support Vector Machine (SVM). To this end, breast Regions of Interest (ROIs) are automatically detected from digital mammograms via gray-scale enhancement and data cleansing. The ROIs are segmented as labeled multi-sectional patterns using spectral clustering by the means of intensity descriptors relying on the region's histogram and texture descriptors based on the Gray Level Co-occurrence Matrix (GLCM). In the next step, shape and probabilistic features are derived from the segmented sections and given to the Genetic Algorithm (GA) to do the feature selection. The optimal feature vector comprising a fusion of selected shape and probabilistic features is submitted to linear kernel SVM for robust and reliable classification of mass tissues from the non-mass. Linear discrimination analysis (LDA) is also performed to ascertain the significance of the nominated feature space. The classification results of the proposed approach are presented by sensitivity, specificity, and accuracy measures, which are 89.5%, 91.2%, and 90%, respectively.

Evaluating attenuation correction strategies in a dedicated, single-gantry breast PET-tomosynthesis scanner

We are developing a dedicated, combined breast positron emission tomography (PET)-tomosynthesis scanner. Both the PET and digital breast tomosynthesis (DBT) scanners are integrated in a single gantry to provide spatially co-registered 3D PET-tomosynthesis images. The DBT image will be used to identify the breast boundary and breast density to improve the quantitative accuracy of the PET image. This paper explores PET attenuation correction (AC) strategies that can be performed with the combined breast PET-DBT scanner to obtain more accurate, quantitative high-resolution 3D PET images. The PET detector is comprised of a 32 × 32 array of 1.5 × 1.5 × 15 mm3 LYSO crystals. The PET scanner utilizes two detector heads separated by either 9 or 11 cm, with each detector head having a 4 × 2 arrangement of PET detectors. GEANT4 Application for Tomographic Emission simulations were performed using an anthropomorphic breast phantom with heterogeneous attenuation under clinical DBT-compression. FDG-avid lesions, each 5 mm in diameter with 8:1 uptake, were simulated at four locations within the breast. Simulations were performed with a scan time of 2 min. PET AC was performed using the actual breast simulation model as well as DBT reconstructed volumetric images to derive the breast outline. In addition to using the known breast density as defined by the breast model, we also modeled it as uniform patient-independent soft-tissue, and as a uniform patient-specific material derived from breast tissue composition. Measured absolute lesion uptake was used to evaluate the quantitative accuracy of performing AC using the various strategies. This study demonstrates that AC is necessary to obtain a closer estimate of the true lesion uptake and background activity in the breast. The DBT image dataset assists in measuring lesion uptake with low bias by facilitating accurate breast delineation as well as providing accurate information related to the breast tissue composition. While both the uniform soft-tissue and patient-specific material approaches provides a close estimate to the ground truth, <5% bias can be achieved by using a uniform patient-specific material to define the attenuation map.

Automated whole breast segmentation for hand-held ultrasound with position information: Application to breast density estimation

BACKGROUND AND OBJECTIVE

Women with higher breast densities have a relatively higher risk to be diagnosed with breast cancer. Hand-held ultrasound (HHUS) can provide precise screening results and detect masses in dense breasts. However, its lack of position information and automatic extraction of breast area hinder the implementation of density estimation. To facilitate reliable breast density evaluation, this study proposed an upgraded version of our whole-breast ultrasound (WBUS) system, which not only can provide precise position information, but also can extract precise breast area automatically based on deep learning method.

METHODS

WBUS images with probe position information were collected from 117 women. For each case, an automatic breast region segmentation by DeepResUnet was conducted, then fibroglandular tissues were extracted from breast region using fuzzy c-mean (FCM) classifier. Finally, the percentage of breast density and breast area of the DeepResUnet predicted region and the breast region of the ground truth were calculated and compared.

RESULTS

The average and standard deviation of each breast case for DeepResUnet predicted breast region of 10-fold in Accuracy (ACC) was 0.963±0.054. Sensitivity (SENS) was 0.928±0.11. Specificity (SPEC) was 0.967±0.054. Dice coefficient (Dice) was 0.916±0.98. Region intersection over union (IoU) was 0.856±0.134. Significant and very high correlations of breast density, fibroglandular tissue area and breast area (R = 0.843, R= 0.822 and R = 0.984, all p values < 0.001) were found between the ground truth and the result of the proposed method for ultrasound images.

CONCLUSIONS

Breast density, fibroglandular tissue, and breast volume evaluated based on the proposed method and WBUS system have significant correlations with ground truth, indicating that the proposed method and WBUS system has the potential to be an alternative modality for breast screening and density estimation in clinical use.

Fully automatic quantification of fibroglandular tissue and background parenchymal enhancement with accurate implementation for axial and sagittal breast MRI protocols

PURPOSE

To propose and evaluate a fully automated technique for quantification of fibroglandular tissue (FGT) and background parenchymal enhancement (BPE) in breast MRI.

METHODS

We propose a fully automated method, where after preprocessing, FGT is segmented in T1-weighted, nonfat-saturated MRI. Incorporating an anatomy-driven prior probability for FGT and robust texture descriptors against intensity variations, our method effectively addresses major image processing challenges, including wide variations in breast anatomy and FGT appearance among individuals. Our framework then propagates this segmentation to dynamic contrast-enhanced (DCE)-MRI to quantify BPE within the segmented FGT regions. Axial and sagittal image data from 40 cancer-unaffected women were used to evaluate our proposed method vs a manually annotated reference standard.

RESULTS

High spatial correspondence was observed between the automatic and manual FGT segmentation (mean Dice similarity coefficient 81.14%). The FGT and BPE quantifications (denoted FGT% and BPE%) indicated high correlation (Pearson's r = 0.99 for both) between automatic and manual segmentations. Furthermore, the differences between the FGT% and BPE% quantified using automatic and manual segmentations were low (mean differences: -0.66 ± 2.91% for FGT% and -0.17 ± 1.03% for BPE%). When correlated with qualitative clinical BI-RADS ratings, the correlation coefficient for FGT% was still high (Spearman's ρ = 0.92), whereas that for BPE was lower (ρ = 0.65). Our proposed approach also performed significantly better than a previously validated method for sagittal breast MRI.

CONCLUSIONS

Our method demonstrated accurate fully automated quantification of FGT and BPE in both sagittal and axial breast MRI. Our results also suggested the complexity of BPE assessment, demonstrating relatively low correlation between segmentation and clinical rating.

Fully Automated Breast Density Segmentation and Classification Using Deep Learning

Breast density estimation with visual evaluation is still challenging due to low contrast and significant fluctuations in the mammograms’ fatty tissue background. The primary key to breast density classification is to detect the dense tissues in the mammographic images correctly. Many methods have been proposed for breast density estimation; nevertheless, most of them are not fully automated. Besides, they have been badly affected by low signal-to-noise ratio and variability of density in appearance and texture. This study intends to develop a fully automated and digitalized breast tissue segmentation and classification using advanced deep learning techniques. The conditional Generative Adversarial Networks (cGAN) network is applied to segment the dense tissues in mammograms. To have a complete system for breast density classification, we propose a Convolutional Neural Network (CNN) to classify mammograms based on the standardization of Breast Imaging-Reporting and Data System (BI-RADS). The classification network is fed by the segmented masks of dense tissues generated by the cGAN network. For screening mammography, 410 images of 115 patients from the INbreast dataset were used. The proposed framework can segment the dense regions with an accuracy, Dice coefficient, Jaccard index of 98%, 88%, and 78%, respectively. Furthermore, we obtained precision, sensitivity, and specificity of 97.85%, 97.85%, and 99.28%, respectively, for breast density classification. This study’s findings are promising and show that the proposed deep learning-based techniques can produce a clinically useful computer-aided tool for breast density analysis by digital mammography.

A Comparison of Regions of Interest in Parenchymal Analysis for Breast Cancer Risk Assessment

Computerized parenchymal analysis has shown potential to be utilized as an imaging biomarker to estimate the risk of breast cancer. Parenchymal analysis of digital mammograms is based on the extraction of computerized measures to build machine learning-based models for the prediction of breast cancer risk. However, the choice of the region of interest (ROI) for feature extraction within the breast remains an open problem. In this work we perform a comparison between five different methods suggested in the literature for automated ROI selection, including the whole breast (WB), the maximum squared (MS), the retro-areolar region (RA), the lattice-based (LB), and the polar-based (PB) selection methods. For the experiments, we built a retrospective dataset of 896 screening mammograms from 224 women (112 cases and 112 healthy controls). The performance of each ROI selection method was measured in terms of the area under the curve (AUC) values. The AUC values varied between 0.55 and 0.79 depending on the method and experimental settings. The best performance on an independent test set was achieved by the MS method (AUC of 0.59, 95% CI: 0.55-0.64). This method is fully-automated and does not require adjusting hyper-parameters. Based on our results, we prompt the use of the MS method for ROI selection in the computerized parenchymal analysis for breast cancer risk assessment.

The Cancer Imaging Phenomics Toolkit (CaPTk): Technical Overview

The purpose of this manuscript is to provide an overview of the technical specifications and architecture of the Cancer imaging Phenomics Toolkit (CaPTk www.cbica.upenn.edu/captk), a cross-platform, open-source, easy-to-use, and extensible software platform for analyzing 2D and 3D images, currently focusing on radiographic scans of brain, breast, and lung cancer. The primary aim of this platform is to enable swift and efficient translation of cutting-edge academic research into clinically useful tools relating to clinical quantification, analysis, predictive modeling, decision-making, and reporting workflow. CaPTk builds upon established open-source software toolkits, such as the Insight Toolkit (ITK) and OpenCV, to bring together advanced computational functionality. This functionality describes specialized, as well as general-purpose, image analysis algorithms developed during active multi-disciplinary collaborative research studies to address real clinical requirements. The target audience of CaPTk consists of both computational scientists and clinical experts. For the former it provides i) an efficient image viewer offering the ability of integrating new algorithms, and ii) a library of readily-available clinically-relevant algorithms, allowing batch-processing of multiple subjects. For the latter it facilitates the use of complex algorithms for clinically-relevant studies through a user-friendly interface, eliminating the prerequisite of a substantial computational background. CaPTk's long-term goal is to provide widely-used technology to make use of advanced quantitative imaging analytics in cancer prediction, diagnosis and prognosis, leading toward a better understanding of the biological mechanisms of cancer development.

Evaluation of LIBRA Software for Fully Automated Mammographic Density Assessment in Breast Cancer Risk Prediction

Background The associations of density measures from the publicly available Laboratory for Individualized Breast Radiodensity Assessment (LIBRA) software with breast cancer have primarily focused on estimates from the contralateral breast at the time of diagnosis. Purpose To evaluate LIBRA measures on mammograms obtained before breast cancer diagnosis and compare their performance to established density measures. Materials and Methods For this retrospective case-control study, full-field digital mammograms in for-processing (raw) and for-presentation (processed) formats were obtained (March 2008 to December 2011) in women who developed breast cancer an average of 2 years later and in age-matched control patients. LIBRA measures included absolute dense area and area percent density (PD) from both image formats. For comparison, dense area and PD were assessed by using the research software (Cumulus), and volumetric PD (VPD) and absolute dense volume were estimated with a commercially available software (Volpara). Density measures were compared by using Spearman correlation coefficients (r), and conditional logistic regression (odds ratios [ORs] and 95% confidence intervals [CIs]) was performed to examine the associations of density measures with breast cancer by adjusting for age and body mass index. Results Evaluated were 437 women diagnosed with breast cancer (median age, 62 years ± 17 [standard deviation]) and 1225 matched control patients (median age, 61 years ± 16). LIBRA PD showed strong correlations with Cumulus PD (r = 0.77-0.84) and Volpara VPD (r = 0.85-0.90) (P < .001 for both). For LIBRA, the strongest breast cancer association was observed for PD from processed images (OR, 1.3; 95% CI: 1.1, 1.5), although the PD association from raw images was not significantly different (OR, 1.2; 95% CI: 1.1, 1.4; P = .25). Slightly stronger breast cancer associations were seen for Cumulus PD (OR, 1.5; 95% CI: 1.3, 1.8; processed images; P = .01) and Volpara VPD (OR, 1.4; 95% CI: 1.2, 1.7; raw images; P = .004) compared with LIBRA measures. Conclusion Automated density measures provided by the Laboratory for Individualized Breast Radiodensity Assessment from raw and processed mammograms correlated with established area and volumetric density measures and showed comparable breast cancer associations. © RSNA, 2020 Online supplemental material is available for this article.

Calculation of Radiomic Features to Validate the Textural Realism of Physical Anthropomorphic Phantoms for Digital Mammography

In this paper, radiomic features are used to validate the textural realism of two anthropomorphic phantoms for digital mammography. One phantom was based off a computational breast model; it was 3D printed by CIRS (Computerized Imaging Reference Systems, Inc., Norfolk, VA) under license from the University of Pennsylvania. We investigate how the textural realism of this phantom compares against a phantom derived from an actual patient's mammogram ("Rachel", Gammex 169, Madison, WI). Images of each phantom were acquired at three kV in 1 kV increments using auto-time technique settings. Acquisitions at each technique setting were repeated twice, resulting in six images per phantom. In the raw ("FOR PROCESSING") images, 341 features were calculated; i.e., gray-level histogram, co-occurrence, run length, fractal dimension, Gabor Wavelet, local binary pattern, Laws, and co-occurrence Laws features. Features were also calculated in a negative screening population. For each feature, the middle 95% of the clinical distribution was used to evaluate the textural realism of each phantom. A feature was considered realistic if all six measurements in the phantom were within the middle 95% of the clinical distribution. Otherwise, a feature was considered unrealistic. More features were actually found to be realistic by this definition in the CIRS phantom (305 out of 341 features or 89.44%) than in the phantom derived from a specific patient's mammogram (261 out of 341 features or 76.54%). We conclude that the texture is realistic overall in both phantoms.