Measurement challenge: protocol for international case-control comparison of mammographic measures that predict breast cancer risk.

INTRODUCTION: For women of the same age and body mass index, increased mammographic density is one of the strongest predictors of breast cancer risk. There are multiple methods of measuring mammographic density and other features in a mammogram that could potentially be used in a screening setting to identify and target women at high risk of developing breast cancer. However, it is unclear which measurement method provides the strongest predictor of breast cancer risk. METHODS AND ANALYSIS: The measurement challenge has been established as an international resource to offer a common set of anonymised mammogram images for measurement and analysis. To date, full field digital mammogram images and core data from 1650 cases and 1929 controls from five countries have been collated. The measurement challenge is an ongoing collaboration and we are continuing to expand the resource to include additional image sets across different populations (from contributors) and to compare additional measurement methods (by challengers). The intended use of the measurement challenge resource is for refinement and validation of new and existing mammographic measurement methods. The measurement challenge resource provides a standardised dataset of mammographic images and core data that enables investigators to directly compare methods of measuring mammographic density or other mammographic features in case/control sets of both raw and processed images, for the purposes of the comparing their predictions of breast cancer risk. ETHICS AND DISSEMINATION: Challengers and contributors are required to enter a Research Collaboration Agreement with the University of Melbourne prior to participation in the measurement challenge. The Challenge database of collated data and images are stored in a secure data repository at the University of Melbourne. Ethics approval for the measurement challenge is held at University of Melbourne (HREC ID 0931343.3).

Technical Note: Validation of two methods to determine contact area between breast and compression paddle in mammography.

PURPOSE: To assess the accuracy of two methods of determining the contact area between the compression paddle and the breast in mammography. An accurate method to determine the contact area is essential to accurately calculate the average compression pressure applied by the paddle. METHODS: For a set of 300 breast compressions, we measured the contact areas between breast and paddle, both capacitively using a transparent foil with indium-tin-oxide (ITO) coating attached to the paddle, and retrospectively from the obtained mammograms using image processing software (Volpara Enterprise, algorithm version 1.5.2). A gold standard was obtained from video images of the compressed breast. During each compression, the breast was illuminated from the sides in order to create a dark shadow on the video image where the breast was in contact with the compression paddle. We manually segmented the shadows captured at the time of x-ray exposure and measured their areas. RESULTS: We found a strong correlation between the manual segmentations and the capacitive measurements [r = 0.989, 95% CI (0.987, 0.992)] and between the manual segmentations and the image processing software [r = 0.978, 95% CI (0.972, 0.982)]. Bland-Altman analysis showed a bias of -0.0038 dm2 for the capacitive measurement (SD 0.0658, 95% limits of agreement [-0.1329, 0.1252]) and -0.0035 dm2 for the image processing software [SD 0.0962, 95% limits of agreement (-0.1921, 0.1850)]. CONCLUSIONS: The size of the contact area between the paddle and the breast can be determined accurately and precisely, both in real-time using the capacitive method, and retrospectively using image processing software. This result is beneficial for scientific research, data analysis and quality control systems that depend on one of these two methods for determining the average pressure on the breast during mammographic compression.

Using Volumetric Breast Density to Quantify the Potential Masking Risk of Mammographic Density.

OBJECTIVE: The purposes of this study were to compare BI-RADS density categories with quantitative volumetric breast density (VBD) for the reporting of mammographic sensitivity and to identify which patient factors are most predictive of a diagnosis of interval cancer of the breast versus screen-detected cancer. MATERIALS AND METHODS: This retrospective study included screen-detected cancers (n = 652) and interval cancers (n = 119) identified between January 2009 and December 2012. Multivariate logistic regression analysis was used to determine which patient factors are predictive of a diagnosis of interval cancer. Sensitivity (screen-detected cancer / [screen-detected cancer + interval cancer]) was determined with the BI-RADS 4th edition density categories and an automated equivalent density grade obtained with a proprietary tool. Sensitivity changes within automated density grade categories were investigated by use of quantitative thresholds at the midpoints of each category. RESULTS: In univariate analysis, age, menopausal status, and breast density were associated with a diagnosis of interval cancer. Of these risk factors, breast density was the only independent factor whether it was assessed by visual BI-RADS category (odds ratio, 3.54; 95% CI, 1.55-8.10), automated density grade (odds ratio, 4.68; 95% CI, 2.26-9.67), or VBD (odds ratio, 4.51; 95% CI, 1.92-10.61). Sensitivity decreased consistently across increasing automated density grade categories from fatty to extremely dense (95%, 89%, 83%, 65%) and less so for visual BI-RADS (82%, 90%, 84%, 66%). Further dichotomization with VBD cutoffs showed a striking linear relation between VBD and sensitivity (R2 = 0.959). CONCLUSION: In this study, breast density was the only risk factor significantly associated with a diagnosis of interval cancer versus screen-detected cancer. Quantitative VBD captures the potential masking risk of breast density more precisely than does the widely used visual BI-RADS density classification system.

Oncological image analysis.

Cancer is one of the world's major healthcare challenges and, as such, an important application of medical image analysis. After a brief introduction to cancer, we summarise some of the major developments in oncological image analysis over the past 20 years, but concentrating those in the authors' laboratories, and then outline opportunities and challenges for the next decade.

Impact of errors in recorded compressed breast thickness measurements on volumetric density classification using volpara v1.5.0 software.

PURPOSE: Mammographic density has been demonstrated to predict breast cancer risk. It has been proposed that it could be used for stratifying screening pathways and recommending additional imaging. Volumetric density tools use the recorded compressed breast thickness (CBT) of the breast measured at the x-ray unit in their calculation; however, the accuracy of the recorded thickness can vary. The aim of this study was to investigate whether inaccuracies in recorded CBT impact upon volumetric density classification and to examine whether the current quality control (QC) standard is sufficient for assessing mammographic density. METHODS: Raw data from 52 digital screening mammograms were included in the study. For each image, the clinically recorded CBT was artificially increased and decreased in increments of 1 mm to simulate measurement error, until ±15% from the recorded CBT was reached. New images were created for each 1 mm step in thickness resulting in a total of 974 images which then had volpara density grade (VDG) and volumetric density percentage assigned. RESULTS: A change in VDG was observed in 38.5% (n = 20) of mammograms when applying ±15% error to the recorded CBT and 11.5% (n = 6) was within the QC standard prescribed error of ±5 mm. CONCLUSIONS: The current QC standard of ±5 mm error in recorded CBT creates the potential for error in mammographic density measurement. This may lead to inaccurate classification of mammographic density. The current QC standard for assessing mammographic density should be reconsidered.

Mammographic compression--a need for mechanical standardization.

BACKGROUND: A lack of consistent guidelines regarding mammographic compression has led to wide variation in its technical execution. Breast compression is accomplished by means of a compression paddle, resulting in a certain contact area between the paddle and the breast. This procedure is associated with varying levels of discomfort or pain. On current mammography systems, the only mechanical parameter available in estimating the degree of compression is the physical entity of force (daN). Recently, researchers have suggested that pressure (kPa), resulting from a specific force divided by contact area on a breast, might be a more appropriate parameter for standardization. Software has now become available which enables device-independent cross-comparisons of key mammographic metrics, such as applied compression pressure (force divided by contact area), breast density and radiation dose, between patient populations. PURPOSE: To compare the current compression practice in mammography between different imaging sites in the Netherlands and the United States from a mechanical point of view, and to investigate whether the compression protocols in these countries can be improved by standardization of pressure (kPa) as an objective mechanical parameter. MATERIALS AND METHODS: We retrospectively studied the available parameters of a set of 37,518 mammographic compressions (9188 women) from the Dutch national breast cancer screening programme (NL data set) and of another set of 7171 compressions (1851 women) from a breast imaging centre in Pittsburgh, PA (US data set). Both sets were processed using VolparaAnalytics and VolparaDensity to obtain the applied average force, pressure, breast thickness, breast volume, breast density and average glandular dose (AGD) as a function of the size of the contact area between the breast and the paddle. RESULTS: On average, the forces and pressures applied in the NL data set were significantly higher than in the US data set. The relative standard deviation was larger in the US data set than in the NL data set. Breasts were compressed with a force in the high range of >15 daN for 31.1% and >20 kPa for 12.3% of the NL data set versus, respectively, 1.5% and 1.7% of the US data set. In the low range we encountered compressions with a pressure of <5 daN for 21.1% and <5 kPa for 21.7% of the US data set versus, respectively, 0.05% and 0.6% in the NL data set. Both the average and the standard deviation of the AGD were higher in the US data set. CONCLUSION: (1) Current mammographic breast compression policies lead to a wide range of applied forces and pressures, with large variations both within and between clinical sites. (2) Pressure standardization could decrease variation, improve reproducibility, and reduce the risk of unnecessary pain, unnecessary high radiation doses and inadequate image quality.

Age and ethnic differences in volumetric breast density in new zealand women: a cross-sectional study.

Breast cancer incidence differs by ethnicity in New Zealand (NZ) with Maori (the indigenous people) women having the highest rates followed by Pakeha (people primarily of British/European descent), Pacific and Asian women, who experience the lowest rates. The reasons for these differences are unclear. Breast density, an important risk factor for breast cancer, has not previously been studied here. We used an automated system, Volpara™, to measure breast density volume from the medio-lateral oblique view of digital mammograms, by age (≤50 years and >50 years) and ethnicity (Pakeha/Maori/Pacific/Asian) using routine data from the national screening programme: age; x-ray system and mammography details for 3,091 Pakeha, 716 Maori, 170 Pacific and 662 Asian (total n = 4,239) women. Linear regression of the natural logarithm of absolute and percent density values was used, back-transformed and expressed as the ratio of the geometric means. Covariates were age, x-ray system and, for absolute density, the natural log of the volume of non-dense tissue (a proxy for body mass index). Median age for Pakeha women was 55 years; Maori 53 years; and Pacific and Asian women, 52 years. Compared to Pakeha women (reference), Maori had higher absolute volumetric density (1.09; 95% confidence interval [95% CI] 1.03-1.15) which remained following adjustment (1.06; 95% CI 1.01-1.12) and was stronger for older compared to younger Maori women. Asian women had the greatest risk of high percentage breast density (1.35; 95% CI 1.27-1.43) while Pacific women in both the ≤50 and >50 year age groups (0.78; 95% CI 0.66-0.92 and 0.81; 95% CI 0.71-0.93 respectively) had the lowest percentage breast density compared to Pakeha. As well as expected age differences, we found differential patterns of breast density by ethnicity consistent with ethnic differences seen in breast cancer risk. Breast density may be a contributing factor to NZ's well-known, but poorly explained, inequalities in breast cancer incidence.

Screen-film mammographic density and breast cancer risk: a comparison of the volumetric standard mammogram form and the interactive threshold measurement methods.

BACKGROUND: Mammographic density is a strong risk factor for breast cancer, usually measured by an area-based threshold method that dichotomizes the breast area on a mammogram into dense and nondense regions. Volumetric methods of breast density measurement, such as the fully automated standard mammogram form (SMF) method that estimates the volume of dense and total breast tissue, may provide a more accurate density measurement and improve risk prediction. METHODS: In 2000-2003, a case-control study was conducted of 367 newly confirmed breast cancer cases and 661 age-matched breast cancer-free controls who underwent screen-film mammography at several centers in Toronto, Canada. Conditional logistic regression was used to estimate odds ratios of breast cancer associated with categories of mammographic density, measured with both the threshold and the SMF (version 2.2beta) methods, adjusting for breast cancer risk factors. RESULTS: Median percent density was higher in cases than in controls for the threshold method (31% versus 27%) but not for the SMF method. Higher correlations were observed between SMF and threshold measurements for breast volume/area (Spearman correlation coefficient = 0.95) than for percent density (0.68) or for absolute density (0.36). After adjustment for breast cancer risk factors, odds ratios of breast cancer in the highest compared with the lowest quintile of percent density were 2.19 (95% confidence interval, 1.28-3.72; P(t) <0.01) for the threshold method and 1.27 (95% confidence interval, 0.79-2.04; Pt = 0.32) for the SMF method. CONCLUSION: Threshold percent density is a stronger predictor of breast cancer risk than the SMF version 2.2beta method in digitized images.

Creating individual-specific biomechanical models of the breast for medical image analysis.

RATIONALE AND OBJECTIVES: Anatomically realistic biomechanical models of the breast potentially provide a reliable way of mapping tissue locations across medical images, such as mammograms, magnetic resonance imaging (MRI), and ultrasound. This work presents a new modeling framework that enables us to create biomechanical models of the breast that are customized to the individual. We demonstrate the framework's capabilities by creating models of the left breasts of two volunteers and tracking their deformations across MRIs. MATERIALS AND METHODS: We generate customized finite element models by automatically fitting geometrical models to segmented data from breast MRIs, and characterizing the in vivo mechanical properties (assuming homogeneity) of the breast tissues. For each volunteer, we identified the unloaded configuration by acquiring MRIs of the breast under neutral buoyancy (immersed in water). Such imaging is clearly not practical in the clinical setting; however, these previously unavailable data provide us with important data with which to validate models of breast biomechanics. Internal tissue features were identified in the neutral buoyancy images and tracked to the prone gravity-loaded state using the modeling framework. RESULTS: The models predicted deformations with root-mean-square errors of 4.2 and 3.6 mm in predicting the skin surface of the gravity-loaded state for each volunteer. Internal tissue features were tracked with a mean error of 3.7 and 4.7 mm for each volunteer. CONCLUSIONS: The models capture breast shape and internal deformations across the images with clinically acceptable accuracy. Further refinement of the framework and incorporation of more anatomic detail will make these models useful for breast cancer diagnosis.

Evaluating the effectiveness of using standard mammogram form to predict breast cancer risk: case-control study.

Breast density is a well-known breast cancer risk factor. Most current methods of measuring breast density are area based and subjective. Standard mammogram form (SMF) is a computer program using a volumetric approach to estimate the percent density in the breast. The aim of this study is to evaluate the current implementation of SMF as a predictor of breast cancer risk by comparing it with other widely used density measurement methods. The case-control study comprised 634 cancers with 1,880 age-matched controls combined from the Cambridge and Norwich Breast Screening Programs. Data collection involved assessing the films based both on Wolfe's parenchymal patterns and on visual estimation of percent density and then digitizing the films for computer analysis (interactive threshold technique and SMF). Logistic regression was used to produce odds ratios associated with increasing categories of breast density. Density measures from all four methods were strongly associated with breast cancer risk in the overall population. The stepwise rises in risk associated with increasing density as measured by the threshold method were 1.37 [95% confidence interval (95% CI), 1.03-1.82], 1.80 (95% CI, 1.36-2.37), and 2.45 (95% CI, 1.86-3.23). For each increasing quartile of SMF density measures, the risks were 1.11 (95% CI, 0.85-1.46), 1.31 (95% CI, 1.00-1.71), and 1.92 (95% CI, 1.47-2.51). After the model was adjusted for SMF results, the threshold readings maintained the same strong stepwise increase in density-risk relationship. On the contrary, once the model was adjusted for threshold readings, SMF outcome was no longer related to cancer risk. The available implementation of SMF is not a better cancer risk predictor compared with the thresholding method.

Mammographic density using two computer-based methods in an isoflavone trial.

OBJECTIVES: Mammographic density is a useful biomarker of breast cancer risk. Computer-based methods can provide continuous data suitable for analysis. This study aimed to compare a semi-automated computer-assisted method (Cumulus) and a fully automated volumetric computer method (standard mammogram form (SMF)) for assessing mammographic density using data from a previously conducted randomised placebo-controlled trial of an isoflavone supplement. METHODS: Mammograms were obtained from participants in the intervention study. A total of 177 women completed the study. Baseline and follow-up mammograms were digitised and density was estimated using Cumulus (read by two readers) and SMF. Left-right correlation, changes in density over time, and difference between intervention and control groups were evaluated. Changes of density over time, and changes between intervention group and control group were examined using paired t-test and Student's t-test, respectively. RESULTS: Inter-reader correlation coefficient by Cumulus was 0.90 for dense area, and 0.86 for percentage density. Left-right correlation of percent density was lower in SMF than in Cumulus. Among all women, percentage density by Cumulus decreased significantly over time, but no change was seen for SMF percentage density. The intervention group showed marginally significant greater reduction of percent density by Cumulus compared to controls (p=0.04), but the difference became weak after adjustment for baseline percent density (p=0.06). No other measurement demonstrated significant difference between intervention and control groups. CONCLUSIONS: This comparison suggests that slightly different conclusions could be drawn from different methods used to assess breast density. The development of a more robust fully automated method is awaited.

Does a short cessation of HRT decrease mammographic density?

BACKGROUND: Hormone replacement therapy (HRT) is known to increase breast density, thus decreasing the sensitivity of cancer screening by mammography. Some authors recommend short cessation of HRT before mammography, but evidence showing the effect of such short cessation is limited. The purpose of this study is to examine whether a short cessation of HRT changes mammographic density. METHODS: Forty-eight women taking HRT agreed to have mammograms taken before and after stopping HRT for 4 weeks. Mammographic density was measured by Wolfe's four-point classification, six-categorical visual scale and two different computer methods (interactive-thresholding and SMF). Density values of mammography before and after the cessation of HRT were compared using Wilcoxon signed-rank test for categorical variables and paired t-test for continuous variables. Changes in breast pain and tenderness during mammography, radiation dose, compression force, and breast thickness were also recorded. RESULTS: No significant changes in mammographic density were observed by either visual or computer methods. There were no significant changes in breast pain or in tenderness on mammograms before and after the month's cessation of HRT. Radiographic measurements were not significantly altered by the 4-week cessation of HRT. CONCLUSION: In this screening population, a 4-week cessation of HRT before mammograms did not significantly alter mammographic density.

Towards tracking breast cancer across medical images using subject-specific biomechanical models.

Breast cancer detection, diagnosis and treatment increasingly involves images of the breast taken with different degrees of breast deformation. We introduce a new biomechanical modelling framework for predicting breast deformation and thus aiding the combination of information derived from the various images. In this paper, we focus on MR images of the breast under different loading conditions, and consider methods to map information between the images. We generate subject-specific finite element models of the breast by semi-automatically fitting geometrical models to segmented data from breast MR images, and characterizing the subject-specific mechanical properties of the breast tissues. We identified the unloaded reference configuration of the breast by acquiring MR images of the breast under neutral buoyancy (immersed in water). Such imaging is clearly not practical in the clinical setting, however this previously unavailable data provides us with important data with which to validate models of breast biomechanics, and provides a common configuration with which to refer and interpret all breast images. We demonstrate our modelling framework using a pilot study that was conducted to assess the mechanical performance of a subject-specific homogeneous biomechanical model in predicting deformations of the breast of a volunteer in a prone gravity-loaded configuration. The model captured the gross characteristics of the breast deformation with an RMS error of 4.2 mm in predicting the skin surface of the gravity-loaded shape, which included tissue displacements of over 20 mm. Internal tissue features identified from the MR images were tracked from the reference state to the prone gravity-loaded configuration with a mean error of 3.7 mm. We consider the modelling assumptions and discuss how the framework could be refined in order to further improve the tissue tracking accuracy.

Comparison of a new and existing method of mammographic density measurement: intramethod reliability and associations with known risk factors.

BACKGROUND: Mammographic density is one of the strongest risk factors for breast cancer. It is commonly measured by an interactive threshold method that does not fully use information contained in a mammogram. An alternative fully automated standard mammogram form (SMF) method measures density using a volumetric approach. METHODS: We examined between-breast and between-view agreement, reliability, and associations of breast cancer risk factors with the threshold and SMF measures of breast density on the same set of 1,000 digitized films from 250 women who attended routine breast cancer screening by two-view mammography in 2004 at a London population-based screening center. Data were analyzed using random-effects models on transformed percent density. RESULTS: Median (interquartile range) percent densities were 12.8% (5.0-22.3) and 21.8% (18.4-26.6) in the threshold and SMF methods, respectively. There was no evidence of systematic differences between left-right breasts or between views in either method. Reliability of a single measurement was lower in the SMF than in the threshold method (0.77 versus 0.92 for craniocaudal and 0.68 versus 0.89 for mediolateral oblique views). Increasing body mass index and parity were associated with reduced density in both methods; however, an increase in density with hormone replacement therapy use was found only with the threshold method. CONCLUSION: Established properties of mammographic density were observed for SMF percent density; however, this method had poorer left-right reliability than the threshold method and has yet to be shown to be a predictor of breast cancer risk.

A mammographic image analysis method to detect and measure changes in breast density.

We present an image analysis method that can detect and measure breast density from digitised mammograms. We present initial results on applying our method to characterise breast changes, in particular, changes due to Hormone Replacement Therapy (HRT). It has been established that long-term use of certain hormone replacement therapies can increase the risk of breast cancer, a fact that encourages the notion that objective measures of tissue density can be an important development in breast cancer image analysis. A set of 59 temporal pairs of mammograms of patients undergoing HRT (two images per patient) were used. The clinician's assessment of density changes constituted the ground truth for evaluating the proposed quantitative measures of density change. The measures we developed are based on the Standard Mammogram Form (SMF) representation of interesting tissue and their performance (agreement with the expert's description) is also compared to the "interactive thresholding" method that has been used in the past to characterise mammographic density. The results clearly indicate that present methods for measuring mammographic density fail to characterise temporal changes while the proposed measures have the potential to aid the radiologist in assessing temporal density changes both on a global and a local basis.

A CAD system for the 3D location of lesions in mammograms.

A CAD system for estimating the 3D (three-dimensional) positions of lesions found in two mammographic views is described. The system is an extension of our previous method [Comput. Vis. Image Understand. 83 (2001) 38] which finds corresponding 2D positions in different mammographic views. The method calculates curved epipolar lines by developing a simulation of breast deformation into stereo camera geometry. Using such curved epipolar lines, not only can we determine point correspondences, but can estimate the 3D location of a lesion. In this paper, we first explain the underlying principles and system organisation. The correctness of the 3D positions calculated by the system is examined using a set of breast lesions, which appear both in mammograms and in MRI data. The experimental results demonstrate the clinical promise of the CAD system.