Volumetric Breast Density Measurement Research Articles

Breast density quantification in dual-energy mammography using virtual anthropomorphic phantoms.

Breast density is a significant risk factor for breast cancer and can impact the sensitivity of screening mammography. Area-based breast density measurements may not provide an accurate representation of the tissue distribution, therefore volumetric breast density (VBD) measurements are preferred. Dual-energy mammography enables volumetric measurements without additional assumptions about breast shape. In this work we evaluated the performance of a dual-energy decomposition technique for determining VBD by applying it to virtual anthropomorphic phantoms. The dual-energy decomposition formalism was used to quantify VBD on simulated dual-energy images of anthropomorphic virtual phantoms with known tissue distributions. We simulated 150 phantoms with volumes ranging from 50 to 709mL and VBD ranging from 15% to 60%. Using these results, we validated a correction for the presence of skin and assessed the method's intrinsic bias and variability. As a proof of concept, the method was applied to 14 sets of clinical dual-energy images, and the resulting breast densities were compared to magnetic resonance imaging (MRI) measurements. Virtual phantom VBD measurements exhibited a strong correlation (Pearson's ) with nominal values. The proposed skin correction eliminated the variability due to breast size and reduced the bias in VBD to a constant value of -2%. Disagreement between clinical VBD measurements using MRI and dual-energy mammography was under 10%, and the difference in the distributions was statistically non-significant. VBD measurements in both modalities had a moderate correlation (Spearman's =0.68). Our results in virtual phantoms indicate that the material decomposition method can produce accurate VBD measurements if the presence of a third material (skin) is considered. The results from our proof of concept showed agreement between MRI and dual-energy mammography VBD. Assessment of VBD using dual-energy images could provide complementary information in dual-energy mammography and tomosynthesis examinations.

Associations of age, body mass index, and breast size with mammographic breast density in Korean women

We aimed (a) to investigate the associations between age, body mass index (BMI), and breast size with mammographic density based on the breast imaging reporting and data system (BI-RADS) and volumetric breast density measurement (VBDM) with Volpara, (b) to evaluate the associations of age, BMI, and breast size with fibroglandular tissue volume (FGV), and (c) to demonstrate the association of mammographic density grade with FGV. From April 2012 to May 2012, 1,203 women consecutively underwent mammography, and their breast density was calculated using the density grade and volume determined by Volpara. In total, 427 women were included in this study. The BMI and breast size of the 427 women were determined. The associations between mammographic density and age, BMI, and bra cup size were assessed. In addition, the associations between FGV and age, BMI, bra cup size, and mammographic density were assessed. The mean age of the women was 51 years (range, 27−83). Age was associated with mammographic density based on BI-RADS (<i>P</i><0.0001), and both age and BMI were associated with mammographic density based on Volpara (<i>P</i><0.0001). The mean FGV significantly decreased as age increased (<i>P</i><0.0001) and increased as BMI and bra cup size increased (<i>P</i><0.0001 and <i>P</i>=0.0007, respectively). Age was associated with mammographic density, according to both the BI-RADS and VBDM; however, BMI was only associated with mammographic density based on the VBDM. Larger FGV was associated with younger age, higher BMI, larger bra cup size, and higher mammographic density

Abstract P022: Quantitative measures of breast density and breast cancer risk prediction among black women in a screening population

Abstract Background: Although mammographic density (MD) is a strong predictor of invasive breast cancer, it has been shown to increase the discriminatory ability of existing risk prediction models only slightly. Breast density is assessed visually by radiologists according to the Breast Imaging Reporting and Data System (BI-RADS) criteria, which has been shown to lack reproducibility. Additionally, the racial diversity of women included in the previous studies was limited. Quantitative measures of breast density have been developed that automatically measure density directly from images. Our prior work found that Black women had lower BI-RADS breast density, despite having a greater quantity of dense breast tissue on average compared with white women when quantitative measures were used. The study purpose was to determine if adding quantitative breast density measures improved breast cancer risk prediction for both white and Black women compared to the Breast Cancer Risk Assessment Tool (BCRAT). Methods: A total of 16,942 women (N=6881 white, N=10061 Black) screened with full-field digital mammography (FFDM) or with a combination of FFDM and digital breast tomosynthesis (DBT) at the Hospital of the University of Pennsylvania (HUP) between September 1, 2010 to December 31, 2014 were included. Area breast density measurements including dense area and area percent density were obtained using a fully automated, validated LIBRA software. All patients were followed from the date of first screening mammogram visit until breast cancer diagnosis or end of follow up on December 31, 2019. We used the BCRA R package (v2.1) for the BCRAT (https://dceg.cancer.gov/tools/risk-assessment/bcra) to estimate the expected 5 year absolute risk for breast cancer. We evaluated the area under the curve (AUC) and calibration (observed to expected ratio, O/E) of the following models: BCRAT alone, BCRAT + BI-RADS density, BCRAT + quantitative density measures, and BCRAT + BI-RADS density + quantitative density measures. Results: There were 123 breast cancers among white and 123 breast cancers among Black women. Adding dense area and area percent density to the BCRAT alone or BCRAT plus BI-RADS density did not improve predictive accuracy for white or Black women. AUC remained close to 0.59 for white women and 0.61 for Black women in all models, with no statistically significant differences in AUCs (DeLong Test p value = 0.09). Underprediction was worse in white women than in Black women. Under-prediction of the BCRAT was reduced when adding percent density from [O/E 1.24 vs. O/E 1.17] in white women. Calibration stayed relatively the same (O/E=1.10) for Black women even when adding both quantitative MD measures. Conclusion: Our results suggest that adding quantitative area mammographic density measurements to the BCRAT does not improves breast cancer risk prediction among Black or white women. Given the increasing use of digital breast tomosynthesis (DBT), future studies should examine whether volumetric breast density measures have superior predictive value among Black women. Citation Format: Mattia A. Mahmoud, Anne Marie McCarthy, Despina Kontos, Emily Conant, Jinbo Chen, Sarah Ehsan, Lauren Pantalone, Walter Mankowski. Quantitative measures of breast density and breast cancer risk prediction among black women in a screening population. [abstract]. In: Proceedings of the AACR Special Conference: Precision Prevention, Early Detection, and Interception of Cancer; 2022 Nov 17-19; Austin, TX. Philadelphia (PA): AACR; Can Prev Res 2023;16(1 Suppl): Abstract nr P022.

Changes in Automated Mammographic Breast Density Can Predict Pathological Response After Neoadjuvant Chemotherapy in Breast Cancer.

Mammographic density is an independent risk factor for breast cancer that can change after neoadjuvant chemotherapy (NCT). This study aimed to evaluate percent changes in volumetric breast density (ΔVbd%) before and after NCT measured automatically and determine its value as a predictive marker of pathological response to NCT. A total of 357 patients with breast cancer treated between January 2014 and December 2016 were included. An automated volumetric breast density (Vbd) measurement method was used to calculate Vbd on mammography before and after NCT. Patients were divided into three groups according to ΔVbd%, calculated as follows: Vbd (post-NCT - pre-NCT)/pre-NCT Vbd × 100 (%). The stable, decreased, and increased groups were defined as -20% ≤ ΔVbd% ≤ 20%, ΔVbd% < -20%, and ΔVbd% > 20%, respectively. Pathological complete response (pCR) was considered to be achieved after NCT if there was no evidence of invasive carcinoma in the breast or metastatic tumors in the axillary and regional lymph nodes on surgical pathology. The association between ΔVbd% grouping and pCR was analyzed using univariable and multivariable logistic regression analyses. The interval between the pre-NCT and post-NCT mammograms ranged from 79 to 250 days (median, 170 days). In the multivariable analysis, ΔVbd% grouping (odds ratio for pCR of 0.420 [95% confidence interval, 0.195-0.905; P = 0.027] for the decreased group compared with the stable group), N stage at diagnosis, histologic grade, and breast cancer subtype were significantly associated with pCR. This tendency was more evident in the luminal B-like and triple-negative subtypes. ΔVbd% was associated with pCR in breast cancer after NCT, with the decreased group showing a lower rate of pCR than the stable group. Automated measurement of ΔVbd% may help predict the NCT response and prognosis in breast cancer.

Significance of quantitative evaluation and assessment using automated volumetric breast density measurements software (Volpara Density)

マンモグラフィ検診の有効性に影響を与える一因として高濃度乳房が知られており，本邦でも２０１６年より議論が盛んに行われるようになった。高濃度乳房の問題は，乳癌の検出率が低下するマスキングリスクと，乳癌の発症が高くなることである。そのため，客観性・再現性を担保した乳房構成の評価は重要である。しかしながら，目視による判定では読影者内・読影者間で乳房構成の評価にばらつきが生じることが報告されている。そこで，画像化処理される前の RAW データから３次元的解析する Volpara 社の乳房構成自動解析ソフト（Volpara Density）を用いて，客観的に乳房構成の評価を行う研究を計画した。 【対象と方法】２０２１年２月より２０２２年９月に共同研究施設で施行されたマンモグラフィ２３，４４７回分を対象とした。被検者は延べ８，３５０人であった。すべて女性で，年齢中央値は４９歳であった。各 RAWデータから乳腺濃度（VBD）は数値化（％）され，a；３．５％未満＝脂肪性，b；３．５～７．５％未満＝乳腺散在，c；７．５～１５．５％未満＝不均一高濃度，d；１５．５％以上＝極めて高濃度に評価される。 【結果】VBD の平均は加齢とともに減少することが示された。また，Grade d（極めて高濃度）の割合も，年齢が上がるとともに減少した。 【まとめ】年齢の上昇によって VBD は低下していくことが客観的に明確となり，日本人における乳房構成の基礎データ（年齢別 VBD と Grade d 率）を得ることができた。Volpara Density は世界中で広く普及している診断ツールであり，大規模な研究を本邦においても導入を急ぐべきと考える。

Prediction of breast cancer risk by automated volumetric breast density measurement.

Dense breast (DB) is recognized as a breast cancer (BC) risk factor. Although DB is common in Japanese women, the incidence of BC is lower than in Caucasians. We evaluated whether DB is a risk factor or whether there are other risk factors for BC in Japanese women. We retrospectively analyzed 635 BC patients and 999 controls who received a mammography at our hospital between February 2019 and March 2021. Volumetric breast density percentage (VBD%), breast volume (BV), and fibroglandular volume (FGV) were measured using Volpara™, an automated, three-dimensional image analysis program. A VBD% of 7.5% or higher was classified as DB. The association between the VBD%, BV, and FGV, and BC risk were assessed using logistic regression. Of the BC group and the control group, 77% and 79% had DB. The stratified FGV was positively associated with BC risk (odds ratio: 2.84; 95% confidence interval 1.58-5.12; P < 0.001). No significant association was found between either the VBD% or BV and BC risk. The proportion of Japanese women with DB was high, suggesting that DB might not be significantly associated with BC risk. However, our results also suggested that the FGV may be related to BC risk in Japanese women.

Abstract 1929: Deep-learning-enabled volumetric breast density estimation with digital breast tomosynthesis

Abstract Background: It has been widely established that breast density is an independent breast cancer risk factor. With the increasing utilization of digital breast tomosynthesis (DBT) in breast cancer screening, there is an opportunity to estimate volumetric breast density (VBD) routinely. However, current available methods extrapolate VBD from 2D images acquired with DBT and/or depend on the existence of raw DBT data which is rarely archived by clinical centers due to cost and storage constraints. This study aims to harness deep learning to develop a computational tool for VBD assessment based solely on 3D reconstructed, “for presentation” DBT images. Methods: We retrospectively analyzed 1,080 negative DBT screening exams (09/20/2011 - 11/25/2016) from the Hospital of the University of Pennsylvania (mean age ± SD, 57 ± 11 years; mean BMI ± SD, 28.7 ± 7.1 kg/m2; racial makeup, 41.2% White, 54.2% Black, 4.6% Other), for which both 3D reconstructed and 2D raw DBT images (Selenia Dimensions, Hologic Inc) were available. All available standard views (left and right mediolateral-oblique and cranio-caudal views) were included for each exam, leading to 7,850 DBT views. Corresponding 3D reference-standard tissue segmentations were generated from a previously validated software that uses both 3D reconstructed slices and raw 2D DBT data to provide VBD metrics, shown to be strongly correlated with VBD measures from MRI image volumes. We based our deep learning algorithm on the U-Net architecture within the open-source Generally Nuanced Deep Learning Framework (GaNDLF) and created a 3-label image segmentation task (background, dense tissue, and fatty tissue). Our dataset was randomly split into training (70%), validation (15%) and test (15%) sets, while ensuring that all views of the same DBT exam were assigned to the same set. The performance of our deep learning algorithm against the corresponding reference-standard segmentations was measured in terms of Dice scores (DSC), with 0 signifying no overlap and 1 signifying perfect overlap, overall, as well as separately for each label. Results: After training was complete, our deep learning algorithm achieved a DSC of 0.78 on the validation, as well as on the test set. Our method accurately segmented background from breast tissue (DSC = 0.94) and demonstrated moderate to high performance in segmenting dense and fatty tissue, respectively (DSC = 0.49 and 0.89). Conclusion: Our preliminary analysis suggests that deep learning shows promise in the estimation of VBD using 3D DBT reconstructed, “for presentation” images. Future work involving transfer learning based on ground truth masks by clinical radiologists could further enhance this method’s performance. In view of rapid clinical conversion to DBT screening, such a tool has the potential to enable large retrospective epidemiologic and personalized risk assessment studies of breast density with DBT. Citation Format: Vinayak S. Ahluwalia, Walter Mankowski, Sarthak Pati, Spyridon Bakas, Ari Brooks, Celine M. Vachon, Emily F. Conant, Aimilia Gastounioti, Despina Kontos. Deep-learning-enabled volumetric breast density estimation with digital breast tomosynthesis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1929.

Artificial-intelligence-driven volumetric breast density estimation with digital breast tomosynthesis in a racially diverse screening cohort.

e13538 Background: Breast density is considered a well-established breast cancer risk factor. As quasi-3D, digital breast tomosynthesis (DBT) becomes increasingly utilized for screening, there is an opportunity to routinely estimate volumetric breast density (VBD). However, current methods extrapolate VBD from 2D images acquired with DBT and/or depend on the existence of raw DBT data, which is rarely archived due to cost and storage constraints. Using a racially diverse screening cohort, this study evaluates the potential of deep learning for VBD assessment based solely on 3D reconstructed, “for presentation” DBT images. Methods: We retrospectively analyzed 1,080 negative DBT screening exams obtained between 2011 and 2016 from the Hospital of the University of Pennsylvania (racial makeup, 41.2% White, 54.2% Black, 4.6% Other; mean age ± SD, 57 ± 11 years; mean BMI ± SD, 28.7 ± 7.1 kg/m2), for which both 2D raw and 3D reconstructed DBT images (Selenia Dimensions, Hologic Inc) were available. Corresponding 3D reference-standard tissue segmentations were generated from previously validated software that uses both 3D reconstructed slices and raw 2D DBT data to provide VBD metrics, shown to be strongly correlated with VBD measures from MRI image volumes. We based our deep learning algorithm on the U-Net architecture within the open-source Generally Nuanced Deep Learning Framework (GaNDLF) and created a 3-label image segmentation task (background, dense tissue, and fatty tissue). Our dataset was randomly split into training (70%), validation (15%) and test (15%) sets. We report on the performance of our deep learning algorithm against corresponding reference-standard segmentations for a cranio-caudal (CC) view-only subset. We also stratify our results by the two main racial groups (White and Black). Our evaluation measure was the weighted Dice score (DSC), with 0 signifying no overlap and 1 signifying perfect overlap, overall and separately for each label. Results: Our deep learning algorithm achieved an overall DSC of 0.682 (STD = 0.136). It accurately segmented the three labels of background, fatty tissue, and dense tissue, with DSC scores of 0.995, 0.884, and 0.617, respectively. DSC for White and Black women were 0.688 (STD = 0.127) and 0.680 (STD = 0.146), respectively. Conclusions: Our preliminary analysis suggests that deep learning shows promise in the estimation of VBD using 3D DBT reconstructed, “for presentation” CC view images and does not demonstrate bias among racial groups. Future work involving optimization of performance in other breast views as well as transfer learning based on ground truth masks by clinical radiologists could further enhance this method. In view of rapid clinical conversion to DBT screening, such a tool has the potential to enable large retrospective epidemiological and personalized risk assessment studies of breast density with DBT.

Comparison of breast density assessments between synthesized C-View™ & intelligent 2D™ mammography.

To compare breast density assessments between C-View™ and Intelligent 2D™, different generations of synthesized mammography (SM) from Hologic. In this retrospective study, we identified a subset of females between March 2017 and December 2019 who underwent screening digital breast tomosynthesis (DBT) with C-View followed by DBT with Intelligent 2D. Clinical Breast Imaging Reporting and Database System breast density was obtained along with volumetric breast density measures (including density grade, breast volume, percentage volumetric density, dense volume) using VolparaTM. Differences in density measures by type of synthesized image were calculated using the pairwise t-test or McNemar's test, as appropriate. 67 patients (avg age 62.7; range 40-84) were included with an average of 13.3 months between the two exams. No difference was found in Breast Imaging Reporting and Database System density between the SM reconstructions (p = 0.74). Similarly, there was no difference in VolparaTM mean density grade (p = 0.71), mean breast volume (p = 0.48), mean dense volume (p = 0.43) or mean percent volumetric density (p = 0.12) between the exams. We found no significant differences in clinical and automated breast density assessments between these two versions of SM. Lack of differences in density estimates between the two SM reconstructions is important as density assignment impacts risk stratification and adjunct screening recommendations.

Mammographic Breast Density and Urbanization: Interactions with BMI, Environmental, Lifestyle, and Other Patient Factors.

Mammographic breast density (MBD) is an important imaging biomarker of breast cancer risk, but it has been suggested that increased MBD is not a genuine finding once corrected for age and body mass index (BMI). This study examined the association of various factors, including both residing in and working in the urban setting, with MBD. Questionnaires were completed by 1144 women attending for mammography at the London Breast Institute in 2012–2013. Breast density was assessed with an automated volumetric breast density measurement system (Volpara) and compared with subjective radiologist assessment. Multivariable linear regression was used to model the relationship between MBD and residence in the urban setting as well as working in the urban setting, adjusting for both age and BMI and other menstrual, reproductive, and lifestyle factors. Urban residence was significantly associated with an increasing percent of MBD, but this association became non-significant when adjusted for age and BMI. This was not the case for women who were both residents in the urban setting and still working. Our results suggest that the association between urban women and increased MBD can be partially explained by their lower BMI, but for women still working, there appear to be other contributing factors.

Race May Not Impact Endocrine Therapy–Related Changes in Breast Density

Reduction in breast density may be a biomarker of endocrine therapy (ET) efficacy. Our objective was to assess the impact of race on ET-related changes in volumetric breast density (VBD). This retrospective cohort study assessed longitudinal changes in VBD measures in women with estrogen receptor-positive invasive breast cancer treated with ET. VBD, the ratio of fibroglandular volume (FGV) to breast volume (BV), was measured using Volpara software. Changes in measurements were evaluated using a multivariable linear mixed effects model. Compared with white women (n = 191), black women (n = 107) had higher rates of obesity [mean ± SD body mass index (BMI) 34.5 ± 9.1 kg/m2 vs. 30.6 ± 7.0 kg/m2, P < 0.001] and premenopausal status (32.7% vs. 16.7%, P = 0.002). Age- and BMI-adjusted baseline FGV, BV, and VBD were similar between groups. Modeled longitudinal changes were also similar: During a follow-up of 30.7 ± 15.0 months (mean ± SD), FGV decreased over time in premenopausal women (slope = -0.323 cm3; SE = 0.093; P = 0.001), BV increased overall (slope = 2.475 cm3; SE = 0.483; P < 0.0001), and VBD decreased (premenopausal slope = -0.063%, SE = 0.011; postmenopausal slope = -0.016%, SE = 0.004; P < 0.0001). Race was not significantly associated with these longitudinal changes, nor did race modify the effect of time on these changes. Higher BMI was associated with lower baseline VBD (P < 0.0001). Among premenopausal women, VBD declined more steeply for women with lower BMI (time × BMI, P = 0.0098). Race does not appear to impact ET-related longitudinal changes in VBD. Racial disparities in estrogen receptor-positive breast cancer recurrence and mortality may not be explained by differential declines in breast density due to ET.

Automated volumetric breast density measures: differential change between breasts in women with and without breast cancer

BackgroundGiven that breast cancer and normal dense fibroglandular tissue have similar radiographic attenuation, we examine whether automated volumetric density measures identify a differential change between breasts in women with cancer and compare to healthy controls.MethodsEligible cases (n = 1160) had unilateral invasive breast cancer and bilateral full-field digital mammograms (FFDMs) at two time points: within 2 months and 1–5 years before diagnosis. Controls (n = 2360) were matched to cases on age and date of FFDMs. Dense volume (DV) and volumetric percent density (VPD) for each breast were assessed using Volpara™. Differences in DV and VPD between mammograms (median 3 years apart) were calculated per breast separately for cases and controls and their difference evaluated by using the Wilcoxon signed-rank test. To simulate clinical practice where cancer laterality is unknown, we examined whether the absolute difference between breasts can discriminate cases from controls using area under the ROC curve (AUC) analysis, adjusting for age, BMI, and time.ResultsAmong cases, the VPD and DV between mammograms of the cancerous breast decreased to a lesser degree (− 0.26% and − 2.10 cm3) than the normal breast (− 0.39% and − 2.74 cm3) for a difference of 0.13% (p value < 0.001) and 0.63 cm3 (p = 0.002), respectively. Among controls, the differences between breasts were nearly identical for VPD (− 0.02 [p = 0.92]) and DV (0.05 [p = 0.77]). The AUC for discriminating cases from controls using absolute difference between breasts was 0.54 (95% CI 0.52, 0.56) for VPD and 0.56 (95% CI, 0.54, 0.58) for DV.ConclusionThere is a small relative increase in volumetric density measures over time in the breast with cancer which is not found in the normal breast. However, the magnitude of this difference is small, and this measure alone does not appear to be a good discriminator between women with and without breast cancer.

Volumetric breast density measurement for personalized screening: accuracy, reproducibility, consistency, and agreement with visual assessment.

Assessment of breast density at the point of mammographic examination could lead to optimized breast cancer screening pathways. The onsite breast density information may offer guidance of when to recommend supplemental imaging for women in a screening program. A software application (Insight BD, Siemens Healthcare GmbH) for fast onsite quantification of volumetric breast density is evaluated. The accuracy of the method is assessed using breast tissue equivalent phantom experiments resulting in a mean absolute error of 3.84%. Reproducibility of measurement results is analyzed using 8427 exams in total, comparing for each exam (if available) the densities determined from left and right views, from cranio-caudal and medio-lateral oblique views, from full-field digital mammograms (FFDM) and digital breast tomosynthesis (DBT) data and from two subsequent exams of the same breast. Pearson correlation coefficients of 0.937, 0.926, 0.950, and 0.995 are obtained. Consistency of the results is demonstrated by evaluating the dependency of the breast density on women's age. Furthermore, the agreement between breast density categories computed by the software with those determined visually by 32 radiologists is shown by an overall percentage agreement of 69.5% for FFDM and by 64.6% for DBT data. These results demonstrate that the software delivers accurate, reproducible, and consistent measurements that agree well with the visual assessment of breast density by radiologists.

The combined effect of mammographic texture and density on breast cancer risk: a cohort study

BackgroundTexture patterns have been shown to improve breast cancer risk segregation in addition to area-based mammographic density. The additional value of texture pattern scores on top of volumetric mammographic density measures in a large screening cohort has never been studied.MethodsVolumetric mammographic density and texture pattern scores were assessed automatically for the first available digital mammography (DM) screening examination of 51,400 women (50–75 years of age) participating in the Dutch biennial breast cancer screening program between 2003 and 2011. The texture assessment method was developed in a previous study and validated in the current study. Breast cancer information was obtained from the screening registration system and through linkage with the Netherlands Cancer Registry. All screen-detected breast cancers diagnosed at the first available digital screening examination were excluded. During a median follow-up period of 4.2 (interquartile range (IQR) 2.0–6.2) years, 301 women were diagnosed with breast cancer. The associations between texture pattern scores, volumetric breast density measures and breast cancer risk were determined using Cox proportional hazard analyses. Discriminatory performance was assessed using c-indices.ResultsThe median age of the women at the time of the first available digital mammography examination was 56 years (IQR 51–63). Texture pattern scores were positively associated with breast cancer risk (hazard ratio (HR) 3.16 (95% CI 2.16–4.62) (p value for trend <0.001), for quartile (Q) 4 compared to Q1). The c-index of texture was 0.61 (95% CI 0.57–0.64). Dense volume and percentage dense volume showed positive associations with breast cancer risk (HR 1.85 (95% CI 1.32–2.59) (p value for trend <0.001) and HR 2.17 (95% CI 1.51–3.12) (p value for trend <0.001), respectively, for Q4 compared to Q1). When adding texture measures to models with dense volume or percentage dense volume, c-indices increased from 0.56 (95% CI 0.53–0.59) to 0.62 (95% CI 0.58–0.65) (p < 0.001) and from 0.58 (95% CI 0.54–0.61) to 0.60 (95% CI 0.57–0.63) (p = 0.054), respectively.ConclusionsDeep-learning-based texture pattern scores, measured automatically on digital mammograms, are associated with breast cancer risk, independently of volumetric mammographic density, and augment the capacity to discriminate between future breast cancer and non-breast cancer cases.

Automated Volumetric Mammographic Breast Density Measurements May Underestimate Percent Breast Density for High-density Breasts

The purpose of this study was to evaluate discrepancy in breast composition measurements obtained from mammograms using two commercially available software methods for systematic trends in overestimation or underestimation compared to magnetic resonance-derived measurements. An institutional review board-approved, Health Insurance Portability and Accountability Act-compliant retrospective study was performed to calculate percent breast density (PBD) by quantifying fibroglandular volume and total breast volume derived from magnetic resonance imaging (MRI) segmentation and mammograms using two commercially available software programs (Volpara and Quantra). Consecutive screening MRI exams from a 6-month period with negative or benign findings were used. The most recent mammogram within 9 months was used to derive mean density values from "for processing" images at the per breast level. Bland-Altman statistical analyses were performed to determine the mean discrepancy and the limits of agreement. A total of 110 women with 220 breasts met the study criteria. Overall, PBD was not different between MRI (mean 10%, range 1%-41%) and Volpara (mean 10%, range 3%-29%); a small but significant difference was present in the discrepancy between MRI and Quantra (4.0%, 95% CI: 2.9 to 5.0, P < 0.001). Discrepancy was highest at higher breast densities, with Volpara slightly underestimating and Quantra slightly overestimating PBD compared to MRI. The mean discrepancy for both Volpara and Quantra for total breast volume was not significantly different from MRI (p = 0.89, 0.35, respectively). Volpara tended to underestimate, whereas Quantra tended to overestimate fibroglandular volume, with the highest discrepancy at higher breast volumes. Both Volpara and Quantra tend to underestimate PBD, which is most pronounced at higher densities. PBD can be accurately measured using automated volumetric software programs, but values should not be used interchangeably between vendors.

Mammographic compression in Asian women.

ObjectivesTo investigate: (1) the variability of mammographic compression parameters amongst Asian women; and (2) the effects of reducing compression force on image quality and mean glandular dose (MGD) in Asian women based on phantom study.MethodsWe retrospectively collected 15818 raw digital mammograms from 3772 Asian women aged 35–80 years who underwent screening or diagnostic mammography between Jan 2012 and Dec 2014 at our center. The mammograms were processed using a volumetric breast density (VBD) measurement software (Volpara) to assess compression force, compression pressure, compressed breast thickness (CBT), breast volume, VBD and MGD against breast contact area. The effects of reducing compression force on image quality and MGD were also evaluated based on measurement obtained from 105 Asian women, as well as using the RMI156 Mammographic Accreditation Phantom and polymethyl methacrylate (PMMA) slabs.ResultsCompression force, compression pressure, CBT, breast volume, VBD and MGD correlated significantly with breast contact area (p<0.0001). Compression parameters including compression force, compression pressure, CBT and breast contact area were widely variable between [relative standard deviation (RSD)≥21.0%] and within (p<0.0001) Asian women. The median compression force should be about 8.1 daN compared to the current 12.0 daN. Decreasing compression force from 12.0 daN to 9.0 daN increased CBT by 3.3±1.4 mm, MGD by 6.2–11.0%, and caused no significant effects on image quality (p>0.05).ConclusionsForce-standardized protocol led to widely variable compression parameters in Asian women. Based on phantom study, it is feasible to reduce compression force up to 32.5% with minimal effects on image quality and MGD.

The impact of using weight estimated from mammographic images vs self-reported weight on breast cancer risk calculation.

Personalised breast screening requires assessment of individual risk of breast cancer, of which one contributory factor is weight. Self-reported weight has been used for this purpose, but may be unreliable. We explore the use of volume of fat in the breast, measured from digital mammograms. Volumetric breast density measurements were used to determine the volume of fat in the breasts of 40,431 women taking part in the Predicting Risk Of Cancer At Screening (PROCAS) study. Tyrer-Cuzick risk using self-reported weight was calculated for each woman. Weight was also estimated from the relationship between self-reported weight and breast fat volume in the cohort, and used to re-calculate Tyrer-Cuzick risk. Women were assigned to risk categories according to 10 year risk (below average <2%, average 2-3.49%, above average 3.5-4.99%, moderate 5-7.99%, high ≥8%) and the original and re-calculated Tyrer-Cuzick risks were compared. Of the 716 women diagnosed with breast cancer during the study, 15 (2.1%) moved into a lower risk category, and 37 (5.2%) moved into a higher category when using weight estimated from breast fat volume. Of the 39,715 women without a cancer diagnosis, 1009 (2.5%) moved into a lower risk category, and 1721 (4.3%) into a higher risk category. The majority of changes were between below average and average risk categories (38.5% of those with a cancer diagnosis, and 34.6% of those without). No individual moved more than one risk group. Automated breast fat measures may provide a suitable alternative to self-reported weight for risk assessment in personalized screening.

Volumetric breast density measurement: sensitivity analysis of a relative physics approach

To investigate the sensitivity and robustness of a volumetric breast density (VBD) measurement system to errors in the imaging physics parameters including compressed breast thickness (CBT), tube voltage (kVp), filter thickness, tube current-exposure time product (mAs), detector gain, detector offset and image noise. 3317 raw digital mammograms were processed with Volpara(®) (Matakina Technology Ltd, Wellington, New Zealand) to obtain fibroglandular tissue volume (FGV), breast volume (BV) and VBD. Errors in parameters including CBT, kVp, filter thickness and mAs were simulated by varying them in the Digital Imaging and Communications in Medicine (DICOM) tags of the images up to ±10% of the original values. Errors in detector gain and offset were simulated by varying them in the Volpara configuration file up to ±10% from their default values. For image noise, Gaussian noise was generated and introduced into the original images. Errors in filter thickness, mAs, detector gain and offset had limited effects on FGV, BV and VBD. Significant effects in VBD were observed when CBT, kVp, detector offset and image noise were varied (p < 0.0001). Maximum shifts in the mean (1.2%) and median (1.1%) VBD of the study population occurred when CBT was varied. Volpara was robust to expected clinical variations, with errors in most investigated parameters giving limited changes in results, although extreme variations in CBT and kVp could lead to greater errors. Despite Volpara's robustness, rigorous quality control is essential to keep the parameter errors within reasonable bounds. Volpara appears robust within those bounds, albeit for more advanced applications such as tracking density change over time, it remains to be seen how accurate the measures need to be.

Influence of fat-water separation and spatial resolution on automated volumetric MRI measurements of fibroglandular breast tissue.

The aim of this study was to investigate the influence of fat-water separation and spatial resolution in MRI on the results of automated quantitative measurements of fibroglandular breast tissue (FGT). Ten healthy volunteers (age range, 28-71 years; mean, 39.9 years) were included in this Institutional Review Board-approved prospective study. All measurements were performed on a 1.5-T scanner (Siemens, AvantoFit) using an 18-channel breast coil. The protocols included isotropic (Di) [TR/TE1 /TE2 = 6.00 ms/2.45 ms/2.67 ms; flip angle, 6.0°; 256 slices; matrix, 360 × 360; 1 mm isotropic; field of view, 360°; acquisition time (TA) = 3 min 38 s] and anisotropic (Da) (TR/TE1 /TE2 = 10.00 ms/2.39 ms/4.77 ms; flip angle, 24.9°; 80 slices; matrix 360 × 360; voxel size, 0.7 × 0.7 × 2.0 mm(3) ; field of view, 360°; TA = 1 min 25 s) T1 three-dimensional (3D) fast low-angle shot (FLASH) Dixon sequences, and a T1 3D FLASH sequence with the same resolution (T1 ) without (TR/TE = 11.00 ms/4.76 ms; flip angle, 25.0°; 80 slices; matrix, 360 × 360; voxel size, 0.7 × 0.7 × 2.0 mm(3) ; field of view, 360°; TA = 50 s) and with (TR/TE = 29.00 ms/4.76 ms; flip angle, 25.0°; 80 slices; matrix, 360 × 360; voxel size, 0.7 × 0.7 × 2.0 mm(3) ; field of view, 360°; TA = 2 min 35 s) fat saturation. Repeating volunteer measurements after 20 min and repositioning were used to assess reproducibility. An automated and quantitative volumetric breast density measurement system was used for FGT calculation. FGT with Di, Da and T1 measured 4.6-63.0% (mean, 30.6%), 3.2-65.3% (mean, 32.5%) and 1.7-66.5% (mean, 33.7%), respectively. The highest correlation between different MRI sequences was found with the Di and Da sequences (R(2) = 0.976). Coefficients of variation (CVs) for FGT calculation were higher in T1 (CV = 21.5%) compared with Dixon (Di, CV = 5.1%; Da, CV = 4.2%) sequences. Dixon-type sequences worked well for FGT measurements, even at lower resolution, whereas the conventional T1 -weighted sequence was more sensitive to decreasing resolution. The Dixon fat-water separation technique showed superior repeatability of FGT measurements compared with conventional sequences. A standard dynamic protocol using Dixon fat-water separation is best suited for combined diagnostic purposes and prognostic measurements of FGT. Copyright © 2016 John Wiley & Sons, Ltd.

Vision 20/20: Mammographic breast density and its clinical applications.

Breast density is a strong predictor of the failure of mammography screening to detect breast cancer and is a strong predictor of the risk of developing breast cancer. The many imaging options that are now available for imaging dense breasts show great promise, but there is still the question of determining which women are "dense" and what imaging modality is suitable for individual women. To date, mammographic breast density has been classified according to the Breast Imaging-Reporting and Data System (BI-RADS) categories from visual assessment, but this is known to be very subjective. Despite many research reports, the authors believe there has been a lack of physics-led and evidence-based arguments about what breast density actually is, how it should be measured, and how it should be used. In this paper, the authors attempt to start correcting this situation by reviewing the history of breast density research and the debates generated by the advocacy movement. The authors review the development of breast density estimation from pattern analysis to area-based analysis, and the current automated volumetric breast density (VBD) analysis. This is followed by a discussion on seeking the ground truth of VBD and mapping volumetric methods to BI-RADS density categories. The authors expect great improvement in VBD measurements that will satisfy the needs of radiologists, epidemiologists, surgeons, and physicists. The authors believe that they are now witnessing a paradigm shift toward personalized breast screening, which is going to see many more cancers being detected early, with the use of automated density measurement tools as an important component.