Assessment of the image contrast improvement and dose reduction in mammography with synchrotron radiation compared to standard units

https://doi.org/10.1016/S0168-9002(01)00662-3Get rights and content

Abstract

An objective method was used to evaluate image quality and dose in mammography with synchrotron radiation and to compare them to standard units. It was performed systematically in the energy range of interest for mammography through the evaluation of the contrast and the measurement of the mean glandular dose. Synchrotron radiation measurements were performed at the ESRF and a slit was placed between the test object and the screen-film system in order to reduce scatter. The conventional films were obtained on mammography units with an anti-scatter grid. In a recent paper, it was shown that the use of synchrotron radiation leads to a noticeable improvement of the image quality-dose relationship (Moeckli et al. Phys. Med. Biol. 45(12)3509). The reason of that enhancement is partly due to the monochromaticity of the synchrotron beam and partly due to the use of a slit instead of a grid. The dose reduction with synchrotron radiation can be attributed to a better X-ray total transmission of the slit and the contrast improvement is due to the monochromaticity of the synchrotron beam.

Introduction

In mammography, the beam quality must be optimised with respect to the breast thickness and composition in order to achieve a high degree of image quality with a minimum amount of radiation delivered to the patient. For X-ray tubes, the combination of anode and filtration materials strongly changes the shape of the energy spectrum and therefore the available tissue contrast. To improve the contrast, an anti-scatter grid is placed between the breast and the screen-film system. The grid reduces scattered radiations, but increases patient dose at the same time.

Third generation synchrotron radiation (SR) sources like the European Synchrotron Radiation Facility (ESRF) in Grenoble produce monochromatic X-ray beams with high photon flux. They are, therefore, a powerful tool to study the effect of the beam energy on image quality and dose in mammography. Different studies showed an enhancement of image quality for an equivalent dose compared to standard mammography units [2], [3], [4], [5].

In order to obtain an objective comparison of performances between SR and mammography units, the image contrast and the mean glandular dose (MGD) were determined.

Section snippets

Materials and method

Images were obtained from a test object consisting of a 51 mm thick PMMA block, representative of an average 55 mm thick breast, including a sheet of 200 μm thick 99.999 % pure aluminium for contrast measurements.

The detector used was a Kodak MinR 2190 - MinR 2000 screen-film system.

Synchrotron measurements were performed on the optics beam line (BM05) and on the medical beam line (ID17) at the ESRF. The beam energy was varied from 15 to 22 keV and the beam height was limited by a tungsten slit to

Contrast–MGD relationship

In order to make the assessment of a mammographic system, image quality performances must be compared with the dose. It was already demonstrated that resolution and noise characteristics were the same for SR and standard units [1]. Thus, this paper will only concentrate on the contrast parameter. Fig. 1 presents the contrast versus MGD. In such a representation, the optimum configuration is placed at the top-left of the plot. Fig. 1 clearly shows that the synchrotron set-up used is advantageous

Summary and conclusion

In order to compare the global performances of the different systems studied, image quality must be analysed with regard to dose. In this study, the image quality-dose relationship (contrast versus MGD) demonstrates that the synchrotron set-up used always leads to a better optimisation of the mammographic process.

This enhancement is due to two reasons: (1) The monochromaticity of the beam improves the image contrast. The reason is the absence of higher energies in the spectrum compared to

Acknowledgments

We greatly acknowledge A. Freund for providing us beam time on BM05. The assistance of T. Brochard from ID17 and R. Hustache from BM05 was greatly appreciated for the experimental setup. We gratefully thank all the staff of ID17 and BM05 for their availability. Finally, we are very grateful to F. Bochud, P.-A. Tercier, R. Porta and I. Decka from IRA for their important work during the synchrotron shifts.

References (6)

  • R. Moeckli et al.

    Phys. Med. Biol.

    (2000)
  • E. Burattini et al.

    Radiology

    (1995)
  • R.A. Lewis et al.

    C. R. M. Boggis Synchrotr. Radiat. News

    (1999)
There are more references available in the full text version of this article.
View full text
Copyright © 2001 Elsevier Science B.V. All rights reserved.