The 2022 European Congress of Radiology concluded on July 17. This five-day congress included three oral presentations and two electronic posters summarizing studies that featured data from Volpara Health products reporting on volumetric breast density, breast compression, and mammography image quality. These studies further demonstrated the importance of:
- Volpara TruDensity® quantitative breast density measurements as a consistent, objective estimate
- The value of Volpara TruPGMI® assessment for proper positioning
- Using TruRadDose® and TruPressure® to provide insights into measurable quality metrics for mammography standards.
Oral Presentations
Worldwide mammographic breast density characterization
Dr. Wendelien Sanderink from Radboud University Medical Center, et al.
This study is an international collaboration on the collection and analysis of volumetric breast density (VBD) measures from Volpara ScorecardTM across different populations. Data from 732,376 women across nine countries were included. The aim was to understand whether, at the same compressed breast thickness, VBD varies between screening populations. The results demonstrated that breasts of comparable thickness have similar VBD distributions. Although the population differences were statistically significant, the absolute differences were small enough that for applications such as dosimetry the differences may not be clinically relevant, while for applications such as cancer risk prediction more accurate VBD estimates may be needed.
Bridging the gap in automatic breast density estimation for screening personalization
Riccardo Samperna from Radboud University Medical Center, et al.
This study was inspired by the DENSE trial findings and resulting EUSOBI recommendations that women with extremely dense breasts be offered supplemental screening, which is preferably done by MRI. As such, they aimed to develop a method to automate density classification from breast MRI such that it matched mammographic density categories. Sets of images from 510 women who had both MRI and mammography acquired within intervals less than one month were retrospectively analyzed. Mammographic density estimates derived from a deep learning-based segmentation tool were split into quintiles and compared to Volpara Density GradesTM (VDG) from Volpara ScorecardTM. There was an exact match between mammographic and MRI-derived VDG classes for 74.4% of studies, with about 20% of cases being rated less dense on MRI than by mammography.
Mammographic breast compression practice: vendor, screening and diagnostic mammography differences
Dr. Anika Brameier from University of Münster and University Hospital Münster, et al.
This retrospective study used Volpara Scorecard results from 58,098 mammographic views to compare breast compression practice differences between three sites in Germany; two screening centers and one diagnostic clinic. This dataset represented imaging done using 22 different units, across which they observed substantial variation in compression force. The system type was found to have the greatest influence on radiation dose, where the site with the highest dose also used the greatest force and pressure, with each screening center parameter being significantly different than those of the diagnostic clinic. Overall, the findings suggested that women may have different experiences between screening and diagnostic imaging services. The researchers concluded that there is a need for pressure-based compression to take breast size into account.
Posters
Mammography image quality indicators associated with interval cancers visible on retrospect
Melissa Hill, PhD, et al.
This retrospective study compared Volpara AnalyticsTM-derived image quality indicators between the prior exams of 87 interval cancer cases where the cancer was visible on retrospect (by either subtle or suspicious signs; ICvr) and 2186 screen-detected cancer (SC) cases from the OPTIMAM database. Compared to SC, increased volumetric breast density and lower compression pressure were associated with being diagnosed as an ICvr. Thus, in addition to density-based screening strategies, compression pressure is an important factor to consider as the results suggested that use of adequate pressure could improve screening performance.
Generative Breast Shape 3D Model for Digital Breast Tomosynthesis
Marta Pinto from Radboud University Medical Center, et al.
The authors present results from 236 subjects who had optical scans of one breast in the CC view during a Digital Breast Tomosynthesis (DBT) exam. Volpara research software was used for DBT image analysis to measure breast characteristics and positioning metrics reported clinically in Volpara Analytics. A 3D model of the breast shape was derived using principal component analysis (PCA), whose components were strongly correlated with Volpara measures of breast volume, and nipple and chest wall positions. The results suggest that the PCA-derived model could predict the 3D compressed breast shape to within 0.68 mm. Among other applications, this type of breast shape model could be used to improve DBT reconstructions and breast density estimates.
Volpara Health developed technologies continue to drive research that will improve the quality of breast imaging using trusted software that enables personalized care in the clinic.
