Search

Purpose Purpose of this work is to compare image quality of Synthetic Mammography (SM) and Full Field Digital Mammography (FFDM) in four different Digital Breast Tomosynthesis (DBT) systems by using a semi-anthropomorphic phantom. Methods Four DBT systems (FujiFilm, GE, Hologic, IMS) were characterized by means of the semi-anthropomorphic CIRS-BR3D phantom. The phantom contains clusters of microcalcifications, fibers and masses embedded in a breast equivalent tissue having a heterogeneous background. For each DBT system two images (DBT and FFDM) were acquired applying clinical setup conditions. FFDM acquisitions were repeated in manual mode for each system with a fixed exposure ratio between DBT and FFDM. Comparative analyses were carried out on the contrasts and sizes of micro calcifications, on the contrasts and shapes of masses, and on the noise power spectrum of the background. The FFDM image acquired in clinical setup conditions was considered the gold standard of the metric. Results Considering the micro calcifications, contrasts were generally preserved by SM with respect to FFDM, while the sizes of the smaller details slightly increased, probably due to the blurring caused by the filtration stages of the tomographic reconstruction. Regarding the masses, no globally significant differences were noticeable in the contrasts, and the shapes were generally well preserved as well. Finally, concerning the noise power spectra, significant variations only appeared in the high frequency regime, in accordance with the above-mentioned blurring. Conclusions This preliminary analysis confirms for all systems the criticality of the synthetic image in terms of spatial resolution and noise but does not highlight significant differences in the detection capability of the details of interest. Further investigations will be made considering the DBT+SM combination compared to DBT+FFDM, as well as by extending the present analysis to other mammographic phantoms.

Threshold contrast visibility measurements are used in mammography to assess image quality. The CDMAM contrast detail phantom version 3.4 was used to set the European standards for quality assurance in digital mammography. As such, the phantom has been widely used to assess the threshold contrast of clinical mammography equipment across Ireland. Recently an updated version of the phantom, version 4.0, was released. The CDMAM 4.0 consists of a new layout with more disc diameters and thicknesses than its predecessor. The new version was created with the aim to be more closely adopted to the image quality found in modern mammography systems and to reduce the variability across individual phantoms. This study compares the sensitivity of the two versions of the CDMAM phantom across a range of mammography systems at clinical settings. Contrast detail measurements were taken using the different phantoms on four different makes of digital mammography equipment and images were analysed using CDMAM Analysis v2.1.0 software package. Here we present resuts of a comparison in threshold contrast sensitivity between the two versions of the phantom.

To demonstrate how a PACS integrated dose monitoring system, Gray Detector, can help to monitor client dose in a breast screening program, ultimately facilitating the optimisation of acquisition parameters in order to enhance image quality and minimise the dose BreastCheck, The National Breast Screening Program evaluated the RIS-PACS integrated dose monitoring system Gray Detector. For each projection of a screening examination, the displayed average glandular dose (AGD), compression force, compressed breast thickness, selected automatic exposure control (AEC) mode, kV and mAs were recorded. Preliminary data have been collected for three months which have enabled comparison of a number of mammography system types and comparison between systems of the same type, in terms of dose, compression force and other key characteristic parameters. Patient dose data collected so far confirm the results of our annual dose survey. All the mammography systems were monitored on a daily basis. The daily presentation of data provides a useful QA tool, highlighting potential problems with individual mammography systems and illustrating the dependency of client dose on other clinical parameters such as compressed breast thickness and compression force. This will enable standardisation of the acquisition technique in order to reduce client dose and optimize the image quality. A summary of these preliminary results will be presented. The Gray Detector software provides useful functionality for monitoring key parameters of the breast screening examination including in particular, the AGD which may be used to monitor and improve clinical quality performance in the screening program.

Getting to grips with quantitative image quality measurements for QA of digital mammography (DM) systems Current European Guidelines for the quality assurance of digital mammography systems recommend and outline methods for the quantitative assessment of image quality (IQ). Recommended metrics include the pre-sampled modulation transfer function (MTF), normalized noise power spectrum (NNPS) and the detective quantum efficiency (DQE). The methods described are well established and have been successfully applied to the evaluation of digital mammography systems for QA purposes. At BreastCheck, we have substantial experience in the use of subjective and semi-quantitative test objects such as the CDMAM contrast-detail test object. As an adjunct to these methods, we have integrated quantitative methods to the routine QA programme. We describe the implementation of quantitative image quality metrics and examine the stability of these parameters over time, the ease with which they might be included in a routine QA programme and discuss some of the problems encountered during the implementation. We have demonstrated good comparison between the MTF values obtained from the QA surveys performed on BreastCheck DM systems with results for similar systems published in the literature. We have also found the long-term reproducibility of MTF, NNPS and DQE measurements values (COV∼3%–5%) are as good as quoted in the literature. We expect that the stability of these IQ measurements will facilitate more accurate monitoring of changes in imaging system performance than can be currently achieved using subjective image quality parameters. Consequently, deteriorating IQ performance may be flagged earlier.